agentleFS
Sign inStart free
sabirmgdsabirmgd/terminal-first-goal-loop/skills/access-check/SKILL.md

access-check

Skill2 starsChanged 2 months agoMITView on GitHub

Locate and verify the access path for a system, credential, or operator surface without exposing secret values. Use when work is blocked on login, cloud access, API credentials, cluster context, CI secrets, or tool authority and the task needs read-only proof of what is available.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 8 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill to answer a simple question safely: do we actually have access, and where does that access come from?

    line 8
  2. Distinctive:

    Never print secret values, reversible encodings, cookies, tokens, kubeconfig bodies, or password-manager contents.

    line 14
  3. Distinctive:

    Verify identity and target explicitly.

    line 17
  4. Distinctive:

    Run the smallest read-only proof.

    line 29
  5. Very specific:

    run a whoami or account command.

    line 40
  6. Distinctive:

    call a read-only health or identity endpoint.

    line 43
  7. Distinctive:

    If a command reveals a secret by default, replace it with a metadata or presence check.

    line 46
  8. Distinctive:

    Stop when the access question is answered with enough proof for the task.

    line 68
  9. Distinctive:

    Stop immediately if the next step would print, decode, rotate, copy, create, or deploy a secret.

    line 70

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Access Check

Use this skill to answer a simple question safely: do we actually have access, and where does that access come from?

Read-Only Rules

  1. Stay read-only.
  2. Never print secret values, reversible encodings, cookies, tokens, kubeconfig bodies, or password-manager contents.
  3. Report the authority, not the secret itself.
  4. Verify identity and target explicitly. Do not trust the current shell context.
  5. Separate source authority from runtime delivery.

Workflow

  1. Name the exact surface.
  2. Identify the authority in this order when possible:
    • secret manager or vault
    • platform credential store
    • dedicated local config
    • CI secret inventory
    • repo handoff file only when it is clearly meant for local access
  3. Run the smallest read-only proof.
  4. Report one of three states:
    • available
    • partial
    • unavailable

available means the required read or write capability was actually proved. A public page, an installed CLI, a secret name, or a working login to a different account is not enough.

Safe Proof Examples

  • list authenticated identities without tokens;
  • run a whoami or account command;
  • list project, cluster, repository, or secret names;
  • inspect file permissions and key names without values;
  • call a read-only health or identity endpoint;
  • confirm a browser session reaches the required signed-in surface.

If a command reveals a secret by default, replace it with a metadata or presence check.

Failure Routing

  • CLI exists but authentication is missing: report the missing login or credential source.
  • Authentication works but the target fails: report the account, project, region, tenant, or permission mismatch.
  • Secret authority is healthy but runtime lacks the value: report a delivery problem, not a missing secret.
  • A public URL works but admin access is unknown: report partial.
  • A local handoff exists but freshness is unknown: report the staleness risk.

Output Contract

ACCESS CHECK | surface=<name> | status=<available|partial|unavailable>
SOURCE | type=<vault|ci|config|browser|other> | location=<path-or-name>
IDENTITY | account=<non-sensitive identity or unknown> | target=<resource>
PROOF | command=<read-only check> | result=<short result>
BLOCKER | none or exact missing piece

Stop Conditions

Stop when the access question is answered with enough proof for the task.

Stop immediately if the next step would print, decode, rotate, copy, create, or deploy a secret. Hand that work to a private access-repair or secret-management workflow with separate authority.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free