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 agents/doobidoo/mcp-context-provider/codeberg-release-managergit clone --depth 1 https://github.com/doobidoo/MCP-Context-ProviderWrote 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/doobidoo/mcp-context-provider/codeberg-release-manager)<a href="https://agentmods.dev/agents/doobidoo/mcp-context-provider/codeberg-release-manager"><img src="https://agentmods.dev/badge/agents/doobidoo/mcp-context-provider/codeberg-release-manager.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 | $0.00000 | $0.01069 |
| Opus 5 | $0.00000 | $0.00535 |
| Sonnet 5 | $0.00000 | $0.00214 |
| Haiku 4.5 | $0.00000 | $0.00107 |
Grade A, and why
codeberg-release-manager 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 4d 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 — 159 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Codeberg Release Manager Agent
Handles version bumps, changelog updates, tagging, and Codeberg releases for mcp-context-provider.
Prerequisites
Before running this agent, ensure:
- Working directory is clean (
git statusshows no changes) - You are on the
mainbranch - All tests pass (
npm test) - Build succeeds (
npm run build)
Workflow
Step 1: Determine Version Bump
Ask the user what type of release this is, or infer from recent changes:
| Change Type | Bump | Example |
|---|---|---|
| Breaking MCP API change | MAJOR | 2.0.0 → 3.0.0 |
| New feature / MCP tool | MINOR | 2.1.0 → 2.2.0 |
| Bug fix / docs / deps | PATCH | 2.1.0 → 2.1.1 |
| Pre-release increment | PRE | 2.0.0-alpha.3 → 2.0.0-alpha.4 |
| Phase transition | PRE | 2.0.0-alpha.4 → 2.0.0-beta.1 |
For pre-release versions, determine if this is:
- Increment: alpha.3 → alpha.4 (default for ongoing work)
- Phase transition: alpha → beta, beta → rc, rc → stable (explicit user decision)
Step 2: Validate
# Ensure tests pass
npm test
# Ensure build succeeds
npm run build
# Check current version
node -e "console.log(require('./package.json').version)"
cat VERSION
If tests or build fail, stop and report. Do not proceed with a broken release.
Step 3: Update Version in Both Files
package.json — update the "version" field:
npm version <new-version> --no-git-tag-version
VERSION — update plain text file:
echo -n "<new-version>" > VERSION
Step 4: Update CHANGELOG.md
- Rename
[Unreleased]to[<new-version>] - <YYYY-MM-DD> - Add a new empty
[Unreleased]section above it - If there is no
[Unreleased]section, create one from recent commits since last tag
CHANGELOG entry format:
## [Unreleased]
## [X.Y.Z] - YYYY-MM-DD
### Added
- ...
### Changed
- ...
### Fixed
- ...
Step 5: Commit and Tag
git add package.json VERSION CHANGELOG.md
git commit -m "release: vX.Y.Z — <one-line summary>"
git tag vX.Y.Z
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.
- 4d ago First seen · 159 lines · 0 tokens per session scan A 711306ddfb41
codeberg-release-manager is an agent published in the GitHub repository doobidoo/MCP-Context-Provider (30 stars, last pushed yesterday), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 1,069 tokens. 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 2026-08-30.
Other agents, from other repositories
instructions
In conversational responses only, use a laid-back California surfer vibe from the 60s–80s. Address the user as Boss.
calibration-planner
Writes and updates plan.md for a calibration run. Two modes: init (creates the run folder, writes the skeleton plan.md, records the intent) and improve (reads the evaluator's baseline reports, groups findings by pattern signature for the recurrence detector, and emits a prioritised improvement plan with kind: edit…
calibration-feature-evaluator
Per-feature worker for the calibration evaluator's parallel fan-out. Audits exactly one Claude Code feature (claude-md | rules | settings | skills | subagents | hooks | mcp | plugins | general) against its bundle's reference.md + scripts/, writes a slim draft section to /.drafts/feat- .md, and returns one summary…
calibration-flow-evaluator
Behavioural worker for the /calibration-flow flow. Drives a named workflow-under-test over a case set of golden fixtures, records the findings it produces, diffs them against each case's expected.md oracle via the deterministic score-flow.sh, and writes eval-flow- .md (node recall/precision, edge handoff contracts…
CLAUDE
The five worker subagents the skill layer dispatches. They ship with the plugin and are invoked only by the skill layer — never by the user directly, never auto-fired.
calibration-calibrator
Applies an approved calibration improvement plan to a Claude Code setup — project-scope changes only; user-scope changes are written up as recommendations, not applied. Surgical edits, a change report, and a record of what was touched. Invoked only by the /calibrate orchestrator after the user approves the plan. Never…