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/agentic-swe/agentic-swe/evaluate-workgit clone --depth 1 https://github.com/agentic-swe/agentic-sweWhat 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 | $0.00013 | $0.00331 |
| Opus 5 | $0.00006 | $0.00166 |
| Sonnet 5 | $0.00003 | $0.00066 |
| Haiku 4.5 | $0.00001 | $0.00033 |
Grade A, and why
evaluate-work scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
/evaluate-work
Inspect an existing work item and assess its health.
Prompt
You are evaluating .worklogs/$ARGUMENTS/.
Instructions:
-
Read
.worklogs/<id>/state.json. -
Verify required artifacts exist for the current state using the artifact table in
CLAUDE.md. -
Check
timeout_atfor staleness. -
Report:
- current state
- last transition (from history)
- missing artifacts
- gate status
- budget status (remaining iterations, cost used vs budget)
- loop counters
- risk score and top items
- whether the run is on the lean track or rigorous track
- whether the run is at an escalation point
- whether the work item is stale (past timeout_at)
-
If this check is requested as an active repair step, update the work item files directly.
-
Optional health scoring: Compare
self-review.md(if present) to${CLAUDE_PLUGIN_ROOT}/templates/evaluation-rubric.md— report any dimension still at 1 after rework limits, and whethermetrics-summary.mdexists under the work directory (optional file for cross-run metrics; not required by the state machine). -
Optional playbook signal: If
docs/agentic-swe/PLAYBOOK.mdexists in the repo, mention the most recent entry’s “next time” line when summarizing risks for the same area of work.
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 · 34 lines · 13 tokens per session scan A 04282d55f19b
evaluate-work is a command published in the GitHub repository agentic-swe/agentic-swe (5 stars, last pushed 3mo ago), licensed MIT. It adds 13 tokens to every session and 331 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.