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/jpantsjoha/ai-native-developer-experience/github-managernpx skills add jpantsjoha/ai-native-developer-experience --skill github-managergit clone --depth 1 https://github.com/jpantsjoha/ai-native-developer-experienceWhat 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 | $0.00038 | $0.01261 |
| Opus 5 | $0.00019 | $0.00630 |
| Sonnet 5 | $0.00008 | $0.00252 |
| Haiku 4.5 | $0.00004 | $0.00126 |
Grade A, and why
github-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 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.
How it starts
The opening of the file, as written. The whole thing — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.
GitHub Manager
CI validates what cannot be cheaply verified locally. It is not a remote test runner for every commit.
This skill enforces disciplined, cost-aware GitHub operations. It turns GitHub from a free-for-all into a governed delivery surface: selective automation, traceable issues, consistent labels, and protected mainlines.
When to use
- Setting up or revising a repository's CI/CD workflow
- Investigating unexpected GitHub Actions spend
- Creating or triaging issues and defining acceptance criteria
- Designing a label taxonomy or branch-protection policy
- Preparing a release process or tag strategy
Operating model context
This skill governs the GitHub surfaces that enforce the harness contract at the repository level. It is not general DevOps housekeeping — each procedure below maps to a harness invariant:
- Branch protection + required status checks are the enforcement mechanism for the PR review gate and the exact-candidate binding rule. A CI check that passes on an unprotected branch is a claim; a passing check required by branch rules is evidence.
- CI receipts are delivery evidence. "It worked locally" is not an artefact. A CI run tied to a commit SHA is. Structure your workflow so evidence is machine-readable and SHA-bound, not dependent on a contributor's local environment.
- Modifying branch protection, CI pipelines, or billing settings is an R2 action. These changes affect all contributors and shared infrastructure. Classify risk, confirm authority, and record the decision before any write.
Use release-readiness to gate a specific deployment. Use this skill to configure and
audit the repository surfaces that make those gates trustworthy.
Procedure
1. Rightsize CI triggers
- Push to default branches should run lightweight gates only (lint, typecheck, unit tests).
- Pull requests and version tags run the full pipeline, including packaging and cross-platform smoke tests.
- Feature branches do not fire CI on every push; open a PR when ready for validation.
- Path filters skip irrelevant jobs: docs-only changes should not rebuild the extension, code-only changes should not re-render documentation.
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 · 113 lines · 38 tokens per session scan A 8ddcfb3e1c57
github-manager is a skill published in the GitHub repository jpantsjoha/ai-native-developer-experience (11 stars, last pushed 25d ago), licensed Apache-2.0. It adds 38 tokens to every session and 1,261 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 2026-08-30.
Other skills, from other repositories
mstar-dispatch-gates
Morning Star 派发与委派门禁 —— 仅 PM 可增派 subagent、Execute as 与 Delegation、承接方反递归 NEVER 红线、SDD implement 串行派发、SDD 路径 plan QC 强制 tri-review（N=3）、inline 单席 QC 例外、Assignment 文案≠派发、未齐不发、invoke 角色字段必填（漏写=静默 generic 回退=派发未完成）。project-manager 派发时必读；leaf 动手前必读反递归。worktree 见 mstar-branch-worktree；SDD 见 mstar-sdd；宿主见 mstar-host。.
hk-scout-wiki
Run a scout wave over outside sources (links, blogs, trending repos) into a progressive-disclosure research wiki in the current project, then distill findings into the project's ideas/backlog file. Generic across projects. Use when the user shares links/blogs to research, asks to "scout", "build a wiki from these", or…
todo
Create and update the Hyper-Waterfall daily task board (mydocs/orders/yyyymmdd.md). Apply milestone table format, status update rules, and backlog section rules. When task-start, task-stage-report, task-final-report, or pr-merge-cleanup updates the daily task board, it follows this SKILL's format.
fable-audit
Audit the fable roadmap (docs/fableroadmap/) — verify claimed progress with proof, write a short report, update RESUME state + BACKLOG rank. Use at wave close or whenever the user asks how roadmap development is going. Triggers on - 'fable audit', 'roadmap audit', 'how is the roadmap going', 'audit progress', or…
sprint
Autonomous multi-agent sprint: Planner decomposes spec → plan checkpoint (user approves, edits, or replans before anything is implemented; auto flag skips it) → Generators implement in parallel via a Claude Code dynamic workflow (Agent-tool fallback when workflows are unavailable) → Evaluator verifies against…
review
L1 internal review with auto-detection, design comparison, and Gate escalation path.