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/askwigconsulting/cohort/project-setupgit clone --depth 1 https://github.com/askwigconsulting/cohortWhat 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 | $0.00019 | $0.00511 |
| Opus 5 | $0.00010 | $0.00255 |
| Sonnet 5 | $0.00004 | $0.00102 |
| Haiku 4.5 | $0.00002 | $0.00051 |
Grade A, and why
project-setup 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 yesterday.
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.
What it actually says
Initialize and tailor Cohort for the current repository. You are conducting a
short interview about the project, then driving the cohort CLI. Show every
artifact before it lands; the human approves each one.
1 — Ensure the project is initialized
Run cohort status --json. If there is no project section, run cohort init
first (it scaffolds .cohort/ and wires project memory).
2 — Project context interview
Ask, one question at a time, and keep answers concise:
- Purpose — what is this project and why does it exist?
- Architecture — the major components and how they fit (read the repo first; confirm your understanding rather than asking cold).
- Decisions — any durable decisions already made, and their rationale.
- Glossary — project-specific terms a newcomer would trip over.
Fill the matching stable sections of .cohort/project_context.md (never touch
the managed Recent sessions block). Show the diff and apply their edits.
3 — Tailored specialists
From the interview and the codebase, propose 1–3 project specialists that would genuinely help (e.g. a schema advisor for a data-heavy repo). For each, draft a real body — Role, Advises on with concrete areas (never "edit me"), Boundaries, Escalation. On approval, write the body to a temp file and run:
cohort add-specialist --name <slug> --display-name <Name> --department <Dept> \
--description '<desc>' --body-file <tempfile>
If a specialist would shadow a global roster agent, say so and let the human decide. Do not create specialists the team did not approve.
4 — Close out
Run cohort snapshot to record the session, and remind the team of the loop:
cohort feedback after working with an agent, cohort propose-improvement
when signals accumulate. .cohort/ (minus state/ and compiled/) is
git-tracked — suggest committing it so the context ships with the repo.
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.
- yesterday First seen · 51 lines · 19 tokens per session scan A 4c2ddc17604c
project-setup is a command published in the GitHub repository askwigconsulting/cohort (2 stars, last pushed 25d ago), licensed MIT. It adds 19 tokens to every session and 511 once invoked, about $0.0001 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 2026-08-31.
Other commands, from other repositories
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.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.