Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/martybonacci/specswarm/mentor-initgit clone --depth 1 https://github.com/MartyBonacci/specswarmWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/martybonacci/specswarm/mentor-init)<a href="https://agentmods.dev/commands/martybonacci/specswarm/mentor-init"><img src="https://agentmods.dev/badge/commands/martybonacci/specswarm/mentor-init.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00101 | $0.00929 |
| Opus 5 | $0.00051 | $0.00464 |
| Sonnet 5 | $0.00020 | $0.00186 |
| Haiku 4.5 | $0.00010 | $0.00093 |
Grade A, and why
mentor-init scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SpecSwarm Mentor Init
Sets up the current directory as a mentor directory — the home of a mentor instance that drives a headless builder via /ss:conduct, and never writes the builder repo's application code itself.
The layout this creates (proven in the Custom Cult v3 pilot, ~500 dispatches):
<mentor-dir>/
├── MENTOR-KICKOFF.md # the standing role — SessionStart hook injects it every session
├── process/
│ ├── escalation-boundary.md # what self-resolves vs subagent-verifies vs goes to the human
│ └── PARKING-LOT.md # verbatim idea capture — the human never holds a mental note
├── dispatch/prompts/ # your authored slice prompts live here
├── .specswarm/conduct/
│ ├── config # BUILDER_DIR, MODEL, TIMEOUT, WATCH_PORTS
│ ├── runs/ # every dispatch's audit trail
│ └── locks/ # one-builder-per-tree, mechanically enforced
└── .claude/
├── settings.json # SessionStart hook wired (merged, never clobbered)
└── hooks/mentor-role.sh # cats MENTOR-KICKOFF.md into session context
Steps
-
Resolve inputs. If
project_nameor--builder-dirwas not provided, ask the user (AskUserQuestion or plain prompt). The builder dir must be an existing git repo — typically the project checked out inside this mentor directory (e.g../my-app). If the intended builder repo doesn't exist yet, stop and have the user clone/create it first. -
Run the scaffold (idempotent — existing files are kept, never overwritten):
bash "${CLAUDE_PLUGIN_ROOT}/lib/conduct/scaffold.sh" \
--project-name "<project_name>" \
--builder-dir "<builder-dir>"
- Review with the user, then hand off. Walk them through what was created, then tell them:
- Edit
MENTOR-KICKOFF.mdto taste — it's the live role text, read by the hook on every session start. - Confirm the Standing Authorizations in
process/escalation-boundary.md(especially the push-without-approval checkbox — that's their call, not yours). - Set
WATCH_PORTSin.specswarm/conduct/configif the project runs dev/e2e webservers. - Restart the Claude Code session in this directory so the SessionStart hook fires — from then on, every session opens already knowing it's the mentor.
- First dispatch:
/ss:conduct(it walks the prompt grammar). - Recommended: install the
ss-statusplugin from this marketplace and run/ss-status:installonce — the statusline then shows background-builder state in real time (🔨 building/✓ safe to close), answering "is anything running?" at a glance with no daemon to manage.
- Edit
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 58 lines · 101 tokens per session scan A 0ed9c3ac742f
mentor-init is a command published in the GitHub repository MartyBonacci/specswarm (65 stars, last pushed 1mo ago), licensed MIT. It adds 101 tokens to every session and 929 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
git
Git operations with intelligent commit messages and workflow optimization.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.