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 skills/foliveira/harmonia/implementnpx skills add foliveira/harmonia --skill implementgit clone --depth 1 https://github.com/foliveira/harmoniaWhat 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 | $0.00032 | $0.00500 |
| Opus 5 | $0.00016 | $0.00250 |
| Sonnet 5 | $0.00006 | $0.00100 |
| Haiku 4.5 | $0.00003 | $0.00050 |
Grade A, and why
implement 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
Your working contract is the 4 rules; their digest is injected at session start - read ${CLAUDE_PLUGIN_ROOT}/core/RULES.md in full only if that digest is not in your context.
Read the implement stage from ${CLAUDE_PLUGIN_ROOT}/core/lifecycle.yaml (already in context if the flow runner loaded it - do not re-read) - agents, artifacts, gates, and the red-green loop definition (including max_rounds) are authoritative; do not hardcode them.
- Workspace:
bash ${CLAUDE_PLUGIN_ROOT}/bin/workspace.sh resolve --repo .(later stage: never mints; on ambiguity or no-active-task, surface the script's message and stop). - Criteria gate (tier B):
bash ${CLAUDE_PLUGIN_ROOT}/bin/check-criteria.sh --workspace <ws> --repo .- refuse to start while it fails (Goal-Driven Execution). It writes its receipt. - Red-green loop, at most
max_roundsrounds:- Dispatch the test-engineer: red-first for behavior; cover-first at gaps named by the last gate report (a green-on-arrival test at a gap completes the round - skip the implementer turn).
bash ${CLAUDE_PLUGIN_ROOT}/bin/workspace.sh record-test-hashes --repo .after every test-engineer turn.- Dispatch the implementer to go green; it may not edit tests.
bash ${CLAUDE_PLUGIN_ROOT}/bin/workspace.sh verify-test-hashes --repo .before accepting the round - a violation fails the round and is recorded for the review lead.bash ${CLAUDE_PLUGIN_ROOT}/bin/coverage/gate.sh --repo . --base <workspace base-ref> --workspace <ws>- its report feeds the next round.- Exit when the gate passes; on the cap, exit incomplete and record the disagreement in the workspace - never escape via an exemption marker.
- Producer duties on completion (KTD10): write
boundary.mdanddiff-summary.mdto the workspace.
Pass workspace paths, not prose recaps (R8). Orchestrate only (R9).
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 · 21 lines · 32 tokens per session scan A a655117084f7
implement is a skill published in the GitHub repository foliveira/harmonia (2 stars, last pushed 17d ago), licensed MIT. It adds 32 tokens to every session and 500 once invoked, about $0.0002 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 skills, from other repositories
snip
You are an expert at writing declarative YAML filters for snip, a CLI proxy that reduces LLM token consumption by filtering shell output.
todos
This chat has a shared, live TODO plan — your tasks for the conversation, which the user also edits. Read this skill and reach for the todo tools whenever a request takes more than a couple of steps. It covers the plan model (group = task, items = its steps; loose items are the user's lane), how to work it: propose…
writing-workflow-skills
Use when adding a new workflow skill to pi-thinkrail-workflow, changing an existing workflow skill's role, trigger, handoff, or structure, or checking a workflow skill against the workflow system's rules. Not for authoring general-purpose skills outside this package.
reviewing-changes
Use when a review package asks you to review a plan step's change set (todo.startReview): you are the REVIEWER, not the author. How to judge an agent-written diff, file findings with addreviewcomment, and settle with exactly one reviewverdict.
setting-up-a-project
Use whenever asked to set up, onboard, initialize, or spec a project — the front door when the workspace has no spec graph yet (brand-new or an existing codebase); also seeded by the app's Set-up-project card (/skill:setting-up-a-project). Not for feature work in an already-specced project — use the brainstorming…
writing-specs
Use when a workflow step drafts or revises a spec artifact — a goal-and-requirements, an architecture, or a module SPEC — or when a workflow skill names it at such a step. The shared quality bar for specs — not a workflow, nothing to execute.