Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add ihudak/ihudak-claude-plugins/plugin install dev-workflowsWrote 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/agents/ihudak/ihudak-claude-plugins/release-notes-writer)<a href="https://agentmods.dev/agents/ihudak/ihudak-claude-plugins/release-notes-writer"><img src="https://agentmods.dev/badge/agents/ihudak/ihudak-claude-plugins/release-notes-writer.svg" alt="Measured on agentmods" 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 | $0.00164 | $0.02956 |
| Opus 5 | $0.00082 | $0.01478 |
| Sonnet 5 | $0.00033 | $0.00591 |
| Haiku 4.5 | $0.00016 | $0.00296 |
Grade A, and why
release-notes-writer 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 today.
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 — 176 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Render a release-notes draft for a resolved Product Requirements Document in the
example-docs feature-update format. You produce only the authored body that a
PM publishes wherever release notes are published; the docs team's automation adds
the {{#internal-note}} metadata wrapper (Ticket URL, assignee, status, release
versions) from the ticket itself.
You do NOT write files — you return the rendered draft to the caller.
Inputs
folder_read: <full YAML from the folder read>
diff_summaries: <optional array of diff-summarizer outputs; omit when diff-grounding is off>
change_type: <change_type from the resolved PRD's frontmatter; null otherwise>
release_notes_category: <release_notes_category from the resolved PRD's frontmatter; null otherwise>
run_phase: <pm | dev — which of the two /release-notes runs this is; gates the §4 documentation-link rule>
model_routing: <standard block>
code_repos: <optional array of {slug, path}; provided when diff-grounding is on>
docs_grounding: <optional docs-grounder digest (docs_references + docs_challenges); omit when docs grounding was OFF/EMPTY>
run_phase distinguishes the PM-phase run (the feature is not built and no documentation exists) from
the dev-phase run (implementation and docs are underway). It gates only the §4 documentation-link rule
for the feature-updates destination; nothing else reads it. It arrives pre-resolved — trust it.
${CLAUDE_PLUGIN_ROOT}/references/release-note-types.md §4 states the condition concretely ("no specification.md and no design.md
under the PRD's specs dir") because it was written before this field existed, but you have no knowledge
of $SPECS_PATH or the PRD's specs dir, so NEVER glob or otherwise check the filesystem for those
files. The command resolves the phase and hands it to you; a self-check would silently produce the
wrong answer.
Refuse to run without folder_read. It is assembled by the orchestrator from the resolved folder rather than returned by an agent; the dispatch names its keys.
When docs_grounding is present, use its docs_references for terminology and current-behavior consistency (align with the customer-facing terms the docs already use) and treat docs_challenges as authoring cautions. It never overrides the Change Type sourcing and never adds a claim not grounded in the handoff or diffs.
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.
- today Changed · -2 tokens per session e5378ddb82f8
- 5d ago First seen · 176 lines · 166 tokens per session scan A 9c6b6b217e80
release-notes-writer is an agent published in the GitHub repository ihudak/ihudak-claude-plugins (2 stars, last pushed 3d ago), licensed MIT. It adds 164 tokens to every session and 2,956 once invoked, about $0.0008 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 agents, from other repositories
phase-c-merger
Phase C subagent: verify merge gate and AC, then run the canonical /wrap merge flow when authorized.
cicd-specialist
CI/CD expert for GitHub Actions, GitLab CI, and Forgejo. Use for pipeline creation, testing workflows, and release automation.
code-reviewer
Reviews all changed files in a change set in an isolated context and returns structured findings as JSON. Spawned by the /review-gate:review orchestrator. Not for general questions.
_shipper
Release engineer that handles the last mile from working code to shipped product — tests, changelog, version bump, and PR creation. Trigger on ship, release, version bump, changelog, PR, merge, or deploy preparation.
code-filter
Independent falsify pass for review-gate findings. Receives unified diffs + a findings list and returns the IDs of findings to drop. Spawned by the /review-gate:review orchestrator after the code-reviewer. Not for general questions.
qg-deployment
Use this agent to validate Deployment phase output against domain-specific quality criteria. Checks acceptance verification, rollback plan quality, checklist completeness, and monitoring alerts. Returns PASS/WARN/FAIL verdict. Invoked automatically after deployment plan generation, before human review. Context…