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/bablsoft/accessflow/prep-gh-releasenpx skills add bablsoft/accessflow --skill prep-gh-releasegit clone --depth 1 https://github.com/bablsoft/accessflowWhat 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 | $0.00075 | $0.06094 |
| Opus 5 | $0.00037 | $0.03047 |
| Sonnet 5 | $0.00015 | $0.01219 |
| Haiku 4.5 | $0.00007 | $0.00609 |
Grade A, and why
prep-gh-release 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 — 361 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Prepare an AccessFlow release
You are preparing the AccessFlow repo for a tagged release. CLAUDE.md is the authoritative rulebook — re-read it before editing anything. This skill formalises the pre-release checklist that used to live in maintainers' heads: every roadmap item must be closed and documented and reflected on the marketing site, every admin-SPA screenshot must be regenerated against the current build, and only then may the milestone be flipped to "released".
The skill does not kick off the Release workflow itself — that stays an explicit maintainer action from the Actions tab (see docs/09-deployment.md → Cutting a release). This skill only prepares the repo so the workflow can run cleanly afterwards.
Inputs
The user passes a target semver, in any of these forms:
1.2v1.21.2.0
Normalize to:
vX.Yfor branch / PR naming (chore/release-prep-vX.Y,chore(release-prep): vX.Y).- The bare heading used in
docs/12-roadmap.md— e.g.## v1.2.
If the user omits the version, ask for it once and stop. Do not guess from the roadmap.
Project map
- Roadmap source of truth:
docs/12-roadmap.md. One## vX.Yheading per milestone, with an✅ releasedor🚧 in progressmarker, then bullet lines that reference(AF-NNN)issues. - Docs chapters (any feature must land in at least one of these):
docs/03-data-model.md— entities, columns, enums, indexesdocs/04-api-spec.md— REST + WebSocket specdocs/05-backend.md— proxy engine, workflow, AI, scheduled jobs, MCPdocs/06-frontend.md— pages, routing, storesdocs/07-security.md— auth, RBAC, encryptiondocs/08-notifications.md— channel types, payloads, retriesdocs/09-deployment.md— Docker / Helm / env vars / releasesdocs/13-mcp.md— MCP server
- README at the repo root — user-facing pitch, tech-stack versions, quick-start, project structure.
- Marketing site at
website/— static HTML/CSS/JS, no build step:website/index.html— landing page (pitch, supported DBs, AI providers, auth methods, feature tiles, roadmap section, quick-start, tech stack, docs grid, footer).website/docs/index.html— public operator docs (deployment, configuration entities, RBAC matrix, env vars).website/README.md— content-source map. Authoritative for "which app/docs source feeds which website section". Read it before judging website coverage.website/images/docs/— admin-SPA screenshots (light + dark pairs for most pages; light-only for editor / queries / reviews by precedent).
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 · 361 lines · 75 tokens per session scan A a9b4bfd82e17
prep-gh-release is a skill published in the GitHub repository bablsoft/accessflow (4 stars, last pushed 2d ago), licensed Apache-2.0. It adds 75 tokens to every session and 6,094 once invoked, about $0.0004 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
connect_polaris_catalog
Guides the agent to ask the user for their preferred authentication mode and credentials when they request to connect to the Polaris catalog, rather than using default credentials.
upgrading-golang
Upgrades Go version across the entire Chainloop codebase including source files, Docker images, CI/CD workflows, and documentation. Use when the user mentions upgrading Go, golang version, or updating Go compiler version.
custom-builtin-functions
Create a custom builtin function to be used in the Rego policy engine.
agt-policy-authoring
Create and validate a minimal AGT Copilot CLI policy tailored to the repository being inspected.
catalyst-center-readonly
Query Cisco Catalyst Center read-only — device inventory, site hierarchy, wireless, assurance health, compliance, software images, events. All 514 read-only API operations reachable through 8 grouped dispatchers. Use when asked what Catalyst Center manages, where a device sits, what its health or compliance state is…
aws-security-audit
AWS security auditing — IAM users/roles/policies, CloudTrail API events, security posture analysis. Use when auditing IAM permissions, investigating security incidents, checking MFA compliance, or tracing API activity in CloudTrail.