rickub
bots-garden/mini-mepublic Fork
⑂ 36b4793
Branches
feature/agents-file feature/ask-agent-hooks feature/ask-cli feature/declared-tools feature/display-called-hooks feature/fetch-build-in feature/genkit-v1.13.1 feature/github-mcp-in-sbx feature/go-1.27.3 feature/hooks feature/improve-init feature/interactive-prompt feature/last-question feature/llmman-provider feature/log-errors feature/lsp-builtin feature/mcp-gateway-session feature/mcp-support feature/memory-search feature/mini-mes feature/mm-init feature/open-terminal feature/prompt-flag feature/response-request feature/resume-acp feature/sbx-template-kit feature/search-in-doc feature/selectors feature/skill-discovery feature/structured-output feature/system-one feature/talk-to-another-mm feature/tasks-builtin feature/try-again feature/unload-model feature/wasm-custom-tools feature/yolo feature/zed-skill-selector fix/generation-timeout fix/plan-mode fix/too-long-completion fix/tool-order fix/word-wrap improve-mem-usage main thinking
Tags
v0.0.0 v0.0.1 v0.0.2 v0.0.3 v0.0.4 v0.1.0 v0.1.1 v0.1.2 v0.1.3 v0.1.4 v0.1.5 v0.1.6 v0.1.7 v0.1.8 v0.1.9 v0.2.0 v0.2.1 v0.2.2 v0.2.3 v0.2.4 v0.2.5 v0.2.6 v0.3.0 v0.4.0 v0.5.0 v0.6.0 v0.7.0 v0.8.0 v0.9.0 v1.0.0 View all branches → View all tags →
Commits
⬇ Clone ▾
git clone https://git.rickub.com/bots-garden/mini-me.git
git clone ssh://git@rickub.com/bots-garden/mini-me.git

Host key fingerprint (ed25519): SHA256:iycHnxEyq0Q7uyVpB7JlznP0G7JrTPXLYRcAU5CSLhc — verify it before your first connect.

Download ZIP

🛟 Updated.

k33g committed 2026-10-02T20:46:40+02:00 Browse files
36b4793 parent: fcfbb59
added .tickets/issues/0066-mini-me-as-a-sdk.yaml +13 -0
new file mode 100644
@@ -0,0 +1,13 @@
1+# Managed by IssueSpec. Hand edits are welcome; keep the schema valid.
2+id: 66
3+title: Mini Me as a SDK
4+state: open
5+author:
6+ name: k33g
7+ email: ph.charriere@gmail.com
8+createdAt: 2026-10-02T05:32:33.878Z
9+updatedAt: 2026-10-02T05:32:33.878Z
10+labels: []
11+body: ""
12+tasks: []
13+comments: []
new file mode 100644
@@ -0,0 +1,13 @@
1+# Managed by IssueSpec. Hand edits are welcome; keep the schema valid.
2+id: 66
3+title: Mini Me as a SDK
4+state: open
5+author:
6+ name: k33g
7+ email: ph.charriere@gmail.com
8+createdAt: 2026-10-02T05:32:33.878Z
9+updatedAt: 2026-10-02T05:32:33.878Z
10+labels: []
11+body: ""
12+tasks: []
13+comments: []
added .tickets/issues/0067-enable-disable-reasoning-mode.yaml +13 -0
new file mode 100644
@@ -0,0 +1,13 @@
1+# Managed by IssueSpec. Hand edits are welcome; keep the schema valid.
2+id: 67
3+title: Enable / Disable reasoning mode
4+state: open
5+author:
6+ name: k33g
7+ email: ph.charriere@gmail.com
8+createdAt: 2026-10-02T06:09:42.599Z
9+updatedAt: 2026-10-02T06:09:42.599Z
10+labels: []
11+body: ""
12+tasks: []
13+comments: []
new file mode 100644
@@ -0,0 +1,13 @@
1+# Managed by IssueSpec. Hand edits are welcome; keep the schema valid.
2+id: 67
3+title: Enable / Disable reasoning mode
4+state: open
5+author:
6+ name: k33g
7+ email: ph.charriere@gmail.com
8+createdAt: 2026-10-02T06:09:42.599Z
9+updatedAt: 2026-10-02T06:09:42.599Z
10+labels: []
11+body: ""
12+tasks: []
13+comments: []
added .tickets/issues/0068-recall-the-last-commands-last.yaml +13 -0
new file mode 100644
@@ -0,0 +1,13 @@
1+# Managed by IssueSpec. Hand edits are welcome; keep the schema valid.
2+id: 68
3+title: Recall the last commands (/last)
4+state: open
5+author:
6+ name: k33g
7+ email: ph.charriere@gmail.com
8+createdAt: 2026-10-02T07:01:13.955Z
9+updatedAt: 2026-10-02T07:01:13.955Z
10+labels: []
11+body: ""
12+tasks: []
13+comments: []
new file mode 100644
@@ -0,0 +1,13 @@
1+# Managed by IssueSpec. Hand edits are welcome; keep the schema valid.
2+id: 68
3+title: Recall the last commands (/last)
4+state: open
5+author:
6+ name: k33g
7+ email: ph.charriere@gmail.com
8+createdAt: 2026-10-02T07:01:13.955Z
9+updatedAt: 2026-10-02T07:01:13.955Z
10+labels: []
11+body: ""
12+tasks: []
13+comments: []
added .tickets/issues/0069-create-skills-to-search-some-specific-knowlege-o.yaml +14 -0
new file mode 100644
@@ -0,0 +1,14 @@
1+# Managed by IssueSpec. Hand edits are welcome; keep the schema valid.
2+id: 69
3+title: Create skills to search some specific knowlege on internet
4+state: open
5+author:
6+ name: k33g
7+ email: ph.charriere@gmail.com
8+createdAt: 2026-10-02T09:26:19.729Z
9+updatedAt: 2026-10-02T09:27:01.494Z
10+labels: []
11+body: |
12+ For example Genkit skill with the appropriate links to the documentation (for fetching)
13+tasks: []
14+comments: []
new file mode 100644
@@ -0,0 +1,14 @@
1+# Managed by IssueSpec. Hand edits are welcome; keep the schema valid.
2+id: 69
3+title: Create skills to search some specific knowlege on internet
4+state: open
5+author:
6+ name: k33g
7+ email: ph.charriere@gmail.com
8+createdAt: 2026-10-02T09:26:19.729Z
9+updatedAt: 2026-10-02T09:27:01.494Z
10+labels: []
11+body: |
12+ For example Genkit skill with the appropriate links to the documentation (for fetching)
13+tasks: []
14+comments: []
added .tickets/issues/0070-add-github-gitlab-other-identity-to-the-mini-me.yaml +13 -0
new file mode 100644
@@ -0,0 +1,13 @@
1+# Managed by IssueSpec. Hand edits are welcome; keep the schema valid.
2+id: 70
3+title: Add GitHub / GitLab / Other ... identity to the Mini Me Agent (but hidden)
4+state: open
5+author:
6+ name: k33g
7+ email: ph.charriere@gmail.com
8+createdAt: 2026-10-02T14:22:29.427Z
9+updatedAt: 2026-10-02T14:22:29.427Z
10+labels: []
11+body: ""
12+tasks: []
13+comments: []
new file mode 100644
@@ -0,0 +1,13 @@
1+# Managed by IssueSpec. Hand edits are welcome; keep the schema valid.
2+id: 70
3+title: Add GitHub / GitLab / Other ... identity to the Mini Me Agent (but hidden)
4+state: open
5+author:
6+ name: k33g
7+ email: ph.charriere@gmail.com
8+createdAt: 2026-10-02T14:22:29.427Z
9+updatedAt: 2026-10-02T14:22:29.427Z
10+labels: []
11+body: ""
12+tasks: []
13+comments: []
added agent.llmman.yaml +200 -0
new file mode 100644
@@ -0,0 +1,200 @@
1+# Agent settings — mini-me served by llmman (https://github.com/llmmanorg/llmman)
2+# through its OpenAI-compatible endpoint (/v1/chat/completions).
3+#
4+# llmman pull hf.co/unsloth/Qwen3.5-4B-GGUF:Q4_K_M # optional: the first question pulls it otherwise
5+# llmman serve # 127.0.0.1:17434
6+#
7+# llmman loads the model on the first request (in its own llama-server) and
8+# unloads it after five idle minutes.
9+provider: llmman
10+
11+# The model id exactly as `llmman list` / GET /v1/models spells it.
12+model: huggingface.co/jetbrains/mellum2-12b-a2.5b-instruct-gguf-q4_k_m:Q4_K_M
13+#model: hf.co/ornith-ai/Ornith-1.5-35B-A3B-GGUF:Q4_K_M
14+
15+# `llmman serve`'s default bind; the OpenAI routes sit under /v1.
16+baseUrl: http://127.0.0.1:17434/v1
17+
18+# From a Docker sandbox, the daemon runs on the host.
19+fallback: http://host.docker.internal:17434/v1
20+
21+# Only when the daemon was started with LLMMAN_API_KEYS (mandatory when it
22+# listens beyond loopback); unset otherwise. The key never goes in this file.
23+# apiKeyEnv: LLMMAN_API_KEY
24+
25+# The `bash` tool.
26+# `false` removes it: the model can still read and edit files (editTools) and load skills,
27+# but never runs a command. If you turn it off, reword the system prompt below — it says "You have a bash tool".
28+bashTool: true
29+
30+# The built-in file tools: read_file, write_file, edit_file.
31+editTools: true
32+
33+# The built-in plan tool for managing multi-step plans.
34+# Set to false to disable plan creation and tracking via the `plan` tool.
35+planTool: true
36+
37+# Language servers for the `lsp` tool (definition, references, hover, symbols, diagnostics, rename).
38+# The file extension picks the server. A server missing from the PATH is skipped with a warning.
39+# No server listed = no `lsp` tool. Examples (uncomment what you have installed):
40+# lsp:
41+# servers:
42+# - name: gopls
43+# command: gopls
44+# extensions: [".go"]
45+# - name: typescript
46+# command: typescript-language-server
47+# args: ["--stdio"]
48+# extensions: [".ts", ".tsx", ".js", ".jsx"]
49+# - name: python
50+# command: pyright-langserver
51+# args: ["--stdio"]
52+# extensions: [".py"]
53+# - name: rust
54+# command: rust-analyzer
55+# extensions: [".rs"]
56+
57+# How many tokens the server actually serves.
58+# 0 = ask llmman: /api/ps tells it once the model is loaded, so the window is
59+# learned before the second question. Set it here if you want it from the start
60+# — and to the same value as LLMMAN_CONTEXT_LENGTH if you set that variable.
61+contextWindow: 0
62+
63+# Max number of characters a tool returns to the model.
64+# Context safeguard: beyond that, the output is truncated (beginning + end kept).
65+maxOutput: 16000
66+
67+# Max number of model <-> tools round trips for a single question.
68+# Raise it for tasks that need long chains of commands.
69+maxTurns: 40
70+
71+# Where the `read_skill` tool looks for its markdown procedures.
72+# The path is relative to the CURRENT DIRECTORY: run the agent from demo-llmman/.
73+# skillsDir: ../skills
74+
75+# Persistent sessions, per project.
76+# When enabled, every conversation is saved as one JSON file under `dir`.
77+# `/sessions` lists them, `/resume` continues one; in an editor, "Reload Agent" replays the thread.
78+sessions:
79+ enabled: true
80+ dir: .mm/sessions
81+
82+# How many lines of a command's output are echoed to the terminal.
83+# 0 disables it.
84+previewLines: 20
85+
86+displayCommands: true
87+# Show what a thinking model reasons about before it answers (dim italic in the
88+# terminal, its own style in the window, a thought chunk in an editor).
89+# The reasoning is always READ — it is what keeps the watchdog alive while the
90+# model thinks — this only says whether to show it.
91+showThinking: true
92+
93+
94+# The system prompt: what the agent is, and what it is allowed to do.
95+# This is the knob to play with — it is the shortest path to changing behaviour.
96+system: |
97+ Your name is Bob.
98+ You are a coding agent working in a terminal.
99+ You have a "bash" tool to run shell commands.
100+ Use it to explore files, run tests, inspect the repository, etc.
101+ Chain several commands if needed, then answer clearly in English.
102+
103+ A request often mixes things you answer from yourself ("say hello") with
104+ things only a command can answer ("list the files"). Handle every part, in
105+ the order asked, and run a command for each part that needs one.
106+ Never state the contents of a file, the output of a command, or the state of
107+ the repository unless a command in THIS answer returned it. What you did not
108+ read, you do not know: run the command instead of recalling it.
109+
110+ SKILLS
111+ You have a second tool, `read_skill`. Its description lists the procedures
112+ available for this project — one per kind of task.
113+
114+ Any request to DO something to a Go project is a skill, not a shell command
115+ you invent. Match the request against that list, call `read_skill` FIRST,
116+ before any bash command, and then follow what it says step by step.
117+
118+ FILE EDITING
119+ You have three tools for files: `read_file`, `edit_file` and `write_file`.
120+ They are how a file gets read and changed here: each change is exact,
121+ checked before it is written, and comes back as a diff with line numbers.
122+ bash is for running things — building, testing, listing, searching.
123+
124+ - Read before you write: call `read_file` on the file (numbered=true when
125+ you need line numbers). You cannot target text you have not seen; never
126+ rely on what you think you remember about a file.
127+ - To change an existing file, call `edit_file` with one or more {old, new}
128+ pairs. `old` is copied from the file character for character — same
129+ spaces, same indentation, same line breaks — and appears exactly once:
130+ add the surrounding lines until it is unique. Several pairs are applied
131+ together, against the original file. An empty `new` deletes the text.
132+ - Call `write_file` only to create a file, or to rewrite one entirely and
133+ on purpose. On an existing file it replaces everything, including what
134+ you did not intend to touch.
135+ - Read the diff the tool returns: it says exactly what changed and on which
136+ line. If `edit_file` refuses — text not found, ambiguous, overlapping
137+ edits — read the file again and fix `old`. Do not fall back to
138+ `write_file` to force the change through.
139+ - After editing code, run the narrowest check with bash: the formatter, the
140+ compiler, or the test covering that file.
141+
142+ RULES
143+ - Keep everything the file already does, unless the user asked to remove it.
144+ - Touch only the files the request is about. Do not add tests, files or
145+ features that were not asked for.
146+ - Never run a git command unless the user says git, commit or push.
147+ - Never move, rename or delete a file unless the user asked for it.
148+ - Then answer in English, in a few lines.
149+ - If you don't know how to use a <cli>, run `<cli> --help` (or `<cli> help`)
150+ to understand the options, then run the command.
151+
152+ BACKGROUND JOBS
153+ Never let a command block the answer. Anything that serves, watches or runs
154+ long goes to the background, with BOTH streams redirected and its pid kept:
155+
156+ nohup <command> > /tmp/<job>.log 2>&1 & echo $! > /tmp/<job>.pid
157+
158+ Redirecting only stdout still blocks until the process exits. Read the
159+ `bg-jobs` skill before you wait on, inspect or stop such a job — each has a
160+ rule you cannot guess. Stop every job you started before you finish, and say
161+ which ones you left running.
162+
163+# Generation settings (OpenAI API keys). Low temperature for a coding agent:
164+# we want precise and reproducible answers, not creativity.
165+sampling:
166+ temperature: 0.0
167+ parallel_tool_calls: false
168+ top_p: 0.9
169+ max_tokens: 4096
170+
171+# Silence allowed between two tokens once the answer has started.
172+watchdogTimeout: 30s
173+
174+# Silence allowed while the model reads the prompt, before the first token:
175+# on a laptop a 25k-token prompt can take minutes. llama-server reports its
176+# progress meanwhile, through llmman ("Reading the prompt 12.0k/25.3k tokens"), and each report
177+# restarts this clock, so it only has to cover one batch of the prompt.
178+firstTokenTimeout: 3m
179+
180+# Context compression.
181+# ON here: the agent learns the window from /api/ps, so `threshold` has a number to apply to.
182+context:
183+ enabled: true
184+
185+ # Compress when the history reaches this share of the window.
186+ threshold: 75
187+
188+ # Fallback on the message count, for when the window is unknown.
189+ # One command costs 2 messages (call + response), so 80 is roughly 30 commands of history.
190+ maxMessages: 80
191+
192+ # How many recent questions (with their commands and answers) are kept as-is instead of being summarised.
193+ # The model works on the last few turns, so summarising them would make it redo what it just did.
194+ keepLastTurns: 3
195+
196+ # max_tokens of the summary request. Reasoning is turned off for it on this provider
197+ # (a thinking model would otherwise spend the budget thinking and write nothing).
198+ summaryMaxTokens: 1200
199+
200+ showStats: true
new file mode 100644
@@ -0,0 +1,200 @@
1+# Agent settings — mini-me served by llmman (https://github.com/llmmanorg/llmman)
2+# through its OpenAI-compatible endpoint (/v1/chat/completions).
3+#
4+# llmman pull hf.co/unsloth/Qwen3.5-4B-GGUF:Q4_K_M # optional: the first question pulls it otherwise
5+# llmman serve # 127.0.0.1:17434
6+#
7+# llmman loads the model on the first request (in its own llama-server) and
8+# unloads it after five idle minutes.
9+provider: llmman
10+
11+# The model id exactly as `llmman list` / GET /v1/models spells it.
12+model: huggingface.co/jetbrains/mellum2-12b-a2.5b-instruct-gguf-q4_k_m:Q4_K_M
13+#model: hf.co/ornith-ai/Ornith-1.5-35B-A3B-GGUF:Q4_K_M
14+
15+# `llmman serve`'s default bind; the OpenAI routes sit under /v1.
16+baseUrl: http://127.0.0.1:17434/v1
17+
18+# From a Docker sandbox, the daemon runs on the host.
19+fallback: http://host.docker.internal:17434/v1
20+
21+# Only when the daemon was started with LLMMAN_API_KEYS (mandatory when it
22+# listens beyond loopback); unset otherwise. The key never goes in this file.
23+# apiKeyEnv: LLMMAN_API_KEY
24+
25+# The `bash` tool.
26+# `false` removes it: the model can still read and edit files (editTools) and load skills,
27+# but never runs a command. If you turn it off, reword the system prompt below — it says "You have a bash tool".
28+bashTool: true
29+
30+# The built-in file tools: read_file, write_file, edit_file.
31+editTools: true
32+
33+# The built-in plan tool for managing multi-step plans.
34+# Set to false to disable plan creation and tracking via the `plan` tool.
35+planTool: true
36+
37+# Language servers for the `lsp` tool (definition, references, hover, symbols, diagnostics, rename).
38+# The file extension picks the server. A server missing from the PATH is skipped with a warning.
39+# No server listed = no `lsp` tool. Examples (uncomment what you have installed):
40+# lsp:
41+# servers:
42+# - name: gopls
43+# command: gopls
44+# extensions: [".go"]
45+# - name: typescript
46+# command: typescript-language-server
47+# args: ["--stdio"]
48+# extensions: [".ts", ".tsx", ".js", ".jsx"]
49+# - name: python
50+# command: pyright-langserver
51+# args: ["--stdio"]
52+# extensions: [".py"]
53+# - name: rust
54+# command: rust-analyzer
55+# extensions: [".rs"]
56+
57+# How many tokens the server actually serves.
58+# 0 = ask llmman: /api/ps tells it once the model is loaded, so the window is
59+# learned before the second question. Set it here if you want it from the start
60+# — and to the same value as LLMMAN_CONTEXT_LENGTH if you set that variable.
61+contextWindow: 0
62+
63+# Max number of characters a tool returns to the model.
64+# Context safeguard: beyond that, the output is truncated (beginning + end kept).
65+maxOutput: 16000
66+
67+# Max number of model <-> tools round trips for a single question.
68+# Raise it for tasks that need long chains of commands.
69+maxTurns: 40
70+
71+# Where the `read_skill` tool looks for its markdown procedures.
72+# The path is relative to the CURRENT DIRECTORY: run the agent from demo-llmman/.
73+# skillsDir: ../skills
74+
75+# Persistent sessions, per project.
76+# When enabled, every conversation is saved as one JSON file under `dir`.
77+# `/sessions` lists them, `/resume` continues one; in an editor, "Reload Agent" replays the thread.
78+sessions:
79+ enabled: true
80+ dir: .mm/sessions
81+
82+# How many lines of a command's output are echoed to the terminal.
83+# 0 disables it.
84+previewLines: 20
85+
86+displayCommands: true
87+# Show what a thinking model reasons about before it answers (dim italic in the
88+# terminal, its own style in the window, a thought chunk in an editor).
89+# The reasoning is always READ — it is what keeps the watchdog alive while the
90+# model thinks — this only says whether to show it.
91+showThinking: true
92+
93+
94+# The system prompt: what the agent is, and what it is allowed to do.
95+# This is the knob to play with — it is the shortest path to changing behaviour.
96+system: |
97+ Your name is Bob.
98+ You are a coding agent working in a terminal.
99+ You have a "bash" tool to run shell commands.
100+ Use it to explore files, run tests, inspect the repository, etc.
101+ Chain several commands if needed, then answer clearly in English.
102+
103+ A request often mixes things you answer from yourself ("say hello") with
104+ things only a command can answer ("list the files"). Handle every part, in
105+ the order asked, and run a command for each part that needs one.
106+ Never state the contents of a file, the output of a command, or the state of
107+ the repository unless a command in THIS answer returned it. What you did not
108+ read, you do not know: run the command instead of recalling it.
109+
110+ SKILLS
111+ You have a second tool, `read_skill`. Its description lists the procedures
112+ available for this project — one per kind of task.
113+
114+ Any request to DO something to a Go project is a skill, not a shell command
115+ you invent. Match the request against that list, call `read_skill` FIRST,
116+ before any bash command, and then follow what it says step by step.
117+
118+ FILE EDITING
119+ You have three tools for files: `read_file`, `edit_file` and `write_file`.
120+ They are how a file gets read and changed here: each change is exact,
121+ checked before it is written, and comes back as a diff with line numbers.
122+ bash is for running things — building, testing, listing, searching.
123+
124+ - Read before you write: call `read_file` on the file (numbered=true when
125+ you need line numbers). You cannot target text you have not seen; never
126+ rely on what you think you remember about a file.
127+ - To change an existing file, call `edit_file` with one or more {old, new}
128+ pairs. `old` is copied from the file character for character — same
129+ spaces, same indentation, same line breaks — and appears exactly once:
130+ add the surrounding lines until it is unique. Several pairs are applied
131+ together, against the original file. An empty `new` deletes the text.
132+ - Call `write_file` only to create a file, or to rewrite one entirely and
133+ on purpose. On an existing file it replaces everything, including what
134+ you did not intend to touch.
135+ - Read the diff the tool returns: it says exactly what changed and on which
136+ line. If `edit_file` refuses — text not found, ambiguous, overlapping
137+ edits — read the file again and fix `old`. Do not fall back to
138+ `write_file` to force the change through.
139+ - After editing code, run the narrowest check with bash: the formatter, the
140+ compiler, or the test covering that file.
141+
142+ RULES
143+ - Keep everything the file already does, unless the user asked to remove it.
144+ - Touch only the files the request is about. Do not add tests, files or
145+ features that were not asked for.
146+ - Never run a git command unless the user says git, commit or push.
147+ - Never move, rename or delete a file unless the user asked for it.
148+ - Then answer in English, in a few lines.
149+ - If you don't know how to use a <cli>, run `<cli> --help` (or `<cli> help`)
150+ to understand the options, then run the command.
151+
152+ BACKGROUND JOBS
153+ Never let a command block the answer. Anything that serves, watches or runs
154+ long goes to the background, with BOTH streams redirected and its pid kept:
155+
156+ nohup <command> > /tmp/<job>.log 2>&1 & echo $! > /tmp/<job>.pid
157+
158+ Redirecting only stdout still blocks until the process exits. Read the
159+ `bg-jobs` skill before you wait on, inspect or stop such a job — each has a
160+ rule you cannot guess. Stop every job you started before you finish, and say
161+ which ones you left running.
162+
163+# Generation settings (OpenAI API keys). Low temperature for a coding agent:
164+# we want precise and reproducible answers, not creativity.
165+sampling:
166+ temperature: 0.0
167+ parallel_tool_calls: false
168+ top_p: 0.9
169+ max_tokens: 4096
170+
171+# Silence allowed between two tokens once the answer has started.
172+watchdogTimeout: 30s
173+
174+# Silence allowed while the model reads the prompt, before the first token:
175+# on a laptop a 25k-token prompt can take minutes. llama-server reports its
176+# progress meanwhile, through llmman ("Reading the prompt 12.0k/25.3k tokens"), and each report
177+# restarts this clock, so it only has to cover one batch of the prompt.
178+firstTokenTimeout: 3m
179+
180+# Context compression.
181+# ON here: the agent learns the window from /api/ps, so `threshold` has a number to apply to.
182+context:
183+ enabled: true
184+
185+ # Compress when the history reaches this share of the window.
186+ threshold: 75
187+
188+ # Fallback on the message count, for when the window is unknown.
189+ # One command costs 2 messages (call + response), so 80 is roughly 30 commands of history.
190+ maxMessages: 80
191+
192+ # How many recent questions (with their commands and answers) are kept as-is instead of being summarised.
193+ # The model works on the last few turns, so summarising them would make it redo what it just did.
194+ keepLastTurns: 3
195+
196+ # max_tokens of the summary request. Reasoning is turned off for it on this provider
197+ # (a thinking model would otherwise spend the budget thinking and write nothing).
198+ summaryMaxTokens: 1200
199+
200+ showStats: true