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/jdanigo/hydraia/plangit clone --depth 1 https://github.com/jdanigo/hydraiaWhat 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 | $0.00018 | $0.00294 |
| Opus 5 | $0.00009 | $0.00147 |
| Sonnet 5 | $0.00004 | $0.00059 |
| Haiku 4.5 | $0.00002 | $0.00029 |
Grade A, and why
plan 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
Invoke the hydraia skill but run ONLY Phases 0-3: context (code graph + PDF conversion if needed), think-before-coding gate, design + threat model, and the detailed implementation plan with its self-review loop (max 2 iterations). Then STOP — do not execute.
Produce BOTH artifacts: the written design spec under docs/hydraia/specs/ (goal, chosen vs rejected approaches, code-graph anchors, global constraints, threat model) and the frozen, file-level-detailed plan under docs/hydraia/plans/ (both relative to the resolved artifacts base — the external dir when chosen at the storage gate). Do NOT arm the spec-drive gate and do NOT write feature code — this is the human review point before implementation.
Request: $ARGUMENTS
When finished, record telemetry for this run: printf 'brief\n' > <base>/.run-complete (where <base> is the artifacts dir resolved at the storage gate — docs/hydraia/ by default, or the external dir if chosen). The Stop hook logs this run's real token/model/sub-agent usage to the local dashboard (delta-scoped per session, so it never double-counts). Do not hand-write the numbers.
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 · 13 lines · 18 tokens per session scan A a9688935de3c
plan is a command published in the GitHub repository jdanigo/hydraia (8 stars, last pushed 13d ago), licensed MIT. It adds 18 tokens to every session and 294 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
MIGRATE_DESIGN
Design doc for the migration tool PR. Author: Sol ([email protected]). Co-authored-by: wakesync.
trex.test.integration
Run integration tests against a real PostgreSQL database (provisioned automatically via testcontainers).
diff-script
Compare an MDL script against the project's current state.
toh-help
Display all Toh Framework commands and quick usage guide.
pipeline-undo
Undo a pipeline run's result. With worktree isolation (the current engine), this is clean and low-risk: a run never touches your checkout — its result lives only on a pipeline/ branch (and, for a --push run, on the remote). "Undo" therefore means deleting that branch and its worktree, not reverting your working tree.
learn-story-flow
Learn story-flow concepts with interactive guidance for junior developers.