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/commit-sequencenpx skills add bablsoft/accessflow --skill commit-sequencegit 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 | $0.00096 | $0.01112 |
| Opus 5 | $0.00048 | $0.00556 |
| Sonnet 5 | $0.00019 | $0.00222 |
| Haiku 4.5 | $0.00010 | $0.00111 |
Grade A, and why
commit-sequence 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 yesterday.
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 — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.
commit-sequence
When to use
A working tree with several unrelated or separable changes. If everything is one change, just commit it — this skill exists to avoid the 40-file "misc" commit nobody can review.
Procedure
1. Read the tree
git status --short
git diff --stat
git diff --cached --stat
Note anything untracked that you did not create — never sweep those in. Ask, or leave them.
2. Propose groups
Present a numbered list: group name, conventional-commit subject, and the files. Then stop and let the user adjust before anything is staged.
3. Group by change, not by directory
AccessFlow's real logical units cut across the tree:
| Unit | Files that belong together |
|---|---|
| A backend feature | the module's api/ + internal/ + its tests + its migration (+ .sql.conf) + messages.properties + the six locales |
| An engine change | engines/<id>/** and connectors/<id>/connector.json — never split: a version bump without the re-pinned SHA fails CI and leaves the catalog pointing at a jar whose hash does not match |
| A frontend feature | src/api/<domain>.ts + the page/components + en.json + the six locales + e2e/tests/<flow>.spec.ts |
| A validation change | the backend DTO constraint and the mirroring Form.Item rule — the parity rule makes these one commit by definition |
| A website change | website/**.html + sitemap.xml (+ frontend/src/config/docs.ts when anchors moved) |
| A config knob | the *Properties record + application.yml + the docs/09-deployment.md row |
4. Never split these
- An engine
pom.xmlversion bump from itsconnectors/<id>/connector.jsonpin. - A migration from its
.sql.confsidecar. - An i18n key from its six translations.
- A backend constraint from its frontend rule.
Splitting any of these leaves an intermediate commit that fails CI — which breaks bisect.
5. Commit each group
Stage explicitly by path (never git add -A), then commit with a heredoc so the body survives:
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.
- yesterday First seen · 100 lines · 96 tokens per session scan A a4fc61ed20f4
commit-sequence is a skill published in the GitHub repository bablsoft/accessflow (4 stars, last pushed yesterday), licensed Apache-2.0. It adds 96 tokens to every session and 1,112 once invoked, about $0.0005 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.
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.
agt-policy-authoring
Create and validate a minimal AGT Copilot CLI policy tailored to the repository being inspected.
auditing-aws-s3-bucket-permissions
Systematically audit AWS S3 bucket permissions to identify publicly accessible buckets, overly permissive ACLs, misconfigured bucket policies, and missing encryption settings using AWS CLI, S3audit, and Prowler to enforce least-privilege data access controls.
github-awesome-copilot-git-commit
Generate high-quality, atomic Conventional Commits by analyzing Git changes, recommending logical commit boundaries, validating commit messages, and assisting with PRs, changelogs, and releases.