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/rasputinkaiser/self-improvement-plugin/verifygit clone --depth 1 https://github.com/RasputinKaiser/Self-Improvement-PluginWhat 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 | $0.00023 | $0.00205 |
| Opus 5 | $0.00012 | $0.00102 |
| Sonnet 5 | $0.00005 | $0.00041 |
| Haiku 4.5 | $0.00002 | $0.00020 |
Grade A, and why
verify 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 2d 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.
What it actually says
Verify the harness is still sound after edits:
python3 ${CLAUDE_PLUGIN_ROOT}/scripts/run_tests.py— report pass/fail counts.- For each file touched in this session, run
python3 ${CLAUDE_PLUGIN_ROOT}/scripts/script_smoke.pyagainst it. - If a test FAILS: do not paper over it. Either the test caught a real regression
(restore via
/checkpoint's snapshot) or the test is stale (flag it, don't silently delete it). - If
proactive_drift.pyreports an untested script, offer to dispatch thetest-authoragent to close the gap.
End with a one-line verdict: VERIFY: OK (n/n tests, k scripts smoke-clean) or
VERIFY: FAIL (details above).
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.
- 2d ago First seen · 17 lines · 23 tokens per session scan A a5484ee3920a
verify is a command published in the GitHub repository RasputinKaiser/Self-Improvement-Plugin (6 stars, last pushed 5d ago), licensed MIT. It adds 23 tokens to every session and 205 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
feature
Orchestrate a complete feature through discovery, spec, implementation, and review.
spec
Turn an objective into a spec and implementable tasks.
research
Research a technical or product question.
fic-create-plan
Act as a Senior XP Developer creating detailed implementation plans through interactive, iterative collaboration.
reqforge-greenkeeper
Diagnose and fix ReqForge repository release gate failures.
evolution-engine
Scan feedback and generate evolution proposals for rule/skill upgrades.