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 skills add yfge/agent-harness-skills --skill work-state-and-deliverygit clone --depth 1 https://github.com/yfge/agent-harness-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/yfge/agent-harness-skills/work-state-and-delivery)<a href="https://agentmods.dev/skills/yfge/agent-harness-skills/work-state-and-delivery"><img src="https://agentmods.dev/badge/skills/yfge/agent-harness-skills/work-state-and-delivery/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/yfge/agent-harness-skills/work-state-and-delivery"><img src="https://agentmods.dev/badge/skills/yfge/agent-harness-skills/work-state-and-delivery.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00048 | $0.01015 |
| Opus 5 | $0.00024 | $0.00508 |
| Sonnet 5 | $0.00010 | $0.00203 |
| Haiku 4.5 | $0.00005 | $0.00102 |
Grade A, and why
work-state-and-delivery 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 11d 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.
How it starts
The opening of the file, as written. The whole thing — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Work State And Delivery
Overview
Keep intent, current work state, acceptance criteria, and delivery evidence connected without creating duplicate sources of truth.
This optional governance skill combines planning surfaces and delivery ledgers because they meet at task-to-change traceability. It also owns repository commit-coupling policy, but it should not load for an ordinary commit that already follows clear repository rules. For shared harness terms, see ../../references/harness-patterns.md; when governance surfaces are missing, use references/build-when-missing.md; for delivery, skip, artifact, and commit rules, use references/delivery-and-commit-policy.md.
When To Use
- The user asks where requirements, design decisions, tasks, acceptance criteria, or execution plans should live.
- A repository needs a task board, external-tracker mapping, delivery record, handoff note, or review evidence policy.
- Design docs, task state, ledgers, reviews, and commits duplicate or contradict each other.
- The user asks to design commit coupling, exact-path staging, task-state updates, or delivery-artifact policy for the repository.
Do not use this skill solely for git status, a read-only diff, or a normal commit when the repository already defines those rules.
Inputs Needed
- Current requirement or delivery goal.
- Existing design docs, issue tracker or task file, execution plans, review template, ledger, and commit policy.
- Validation results, artifact references, task IDs, commits, risks, and required audit level.
Execution Order
- First: Map existing design, work-state, execution-plan, ledger, review, and commit-policy surfaces.
- Then: Assign one owner for each kind of truth and define update, coupling, skip, and redaction rules.
- Finally: Output the minimum governance changes, acceptance criteria, and delivery evidence required for the repository profile.
Step-by-Step Process
- Search for design docs,
tasks.md, issue references, execution plans, review templates, ledgers, release notes, and commit guidance. - Map each existing surface by responsibility rather than filename: design explains why and how; work state owns current status and acceptance; an execution plan coordinates complex work; delivery evidence records what changed and what was proved.
- If a required role is missing, bootstrap only the minimum surface from
references/build-when-missing.md; do not add a ledger or repo-local task file when an external system already satisfies the role. - Define update order: design decision first, work-state and acceptance second, execution-plan status third, delivery evidence after validation.
- When a change completes, changes, or invalidates a tracked task, include its task-state update in the same logical commit or authoritative external-tracker update.
- Define exact-path staging, validation evidence, artifact identity, and explicit skip rules in repository guidance; avoid a standalone commit skill for routine execution.
- Keep delivery records concise and evidence-oriented; reference prompts or goals, changes, validation, artifacts, risks, tasks, reviews, and commits without storing private transcripts.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 11d ago First seen · 97 lines · 48 tokens per session scan A fb89e7903a0c
work-state-and-delivery is a skill published in the GitHub repository yfge/agent-harness-skills (10 stars, last pushed 1mo ago), licensed MIT. It adds 48 tokens to every session and 1,015 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
triage
Triage a GitHub repository's open issues by finding exact duplicates, rejecting evidenceably off-base requests, requesting concrete clarification, applying only existing labels, and opening a linked root-cause issue when multiple reports share one underlying invariant failure. Use when a maintainer asks to triage…
to-goal
Turn an approved spec, agent-ready ticket, tracker frontier, or partially implemented ticket into a verifiable execution goal without re-interviewing the user. Use after to-tickets or triage, before starting a fresh implementation session; use --all only for an explicitly requested cross-ticket goal.
handing-off-work
Hands off unfinished work with a closed-loop briefing of state, changed conditions, remaining scope, authority limits, and open evidence. Use when AI-agent, reviewer, verifier, releaser, or resumed-thread work transfers to a new owner. Do not use when the same owner continues uninterrupted with full context.
tracking-deficiencies
Keeps a standing register of known deficiencies so flaky tests, noisy alerts, unowned services, and recurring incidents get aged, owned, and fixed or formally risk-accepted instead of quietly normalized. Use when a known problem will outlive a single change. Do not use for a one-off lesson already closed inside a…
pace-bridge
Use to bridge a confirmed Superpowers/native plan into PACEflow CHG/HOTFIX artifacts, create artifact-writer prompts, and mark the specific plan as synced.
readonly-review
Run a no-network read-only review practice with two explorer steps and one synthesis step.