chronometric-mark-start-7fd199 — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited chronometric-mark-start-7fd199 (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Quick wrapper for the mark_session_start tool on the local NeuroDock chronometric server (neurodock-chronometric). Captures the user's plain-language framing of what they are about to do, so later nudges (hyperfocus warnings, end-of-session reflection) can quote it back unchanged.
Authoritative schema $id: https://schemas.neurodock.org/mcp-chronometric/v0.1.0/mark_session_start.schema.json (source).
finish the RFC reply now").
session against a stated intent.
Calls mcp__neurodock-chronometric__mark_session_start with a single intent string. The server:
session_id.started_at (ISO 8601 with offset).intent verbatim — no paraphrase, no summarisation.includes auto_closed_prior_session with the prior id and closed_at.
Input shape:
{
"intent": "finish draft RFC reply"
}Constraints:
intent is required, 1–500 chars, plain language.before calling, or accept the warning.
Example response:
{
"session_id": "6f9f4f5e-3b1c-4e2a-9f7d-2b3c4d5e6f70",
"started_at": "2026-05-23T09:14:22+01:00",
"intent": "finish draft RFC reply",
"auto_closed_prior_session": null
}The session_id is opaque. Callers do NOT need to track it across calls — mark_session_end always closes the most-recent open session for the user by design (see ADR 0001 — avoids the LLM having to maintain state).
The id is useful for:
source: "session://<id>").INTENT_REQUIRED — empty/missing intent. Ask the user to state one.INTENT_TOO_LONG — over 500 chars. Truncate or split.SESSION_ALREADY_OPEN — only when the server is configured to rejectoverlaps. The default policy auto-closes.
is upstream; this server analyses what it is given.
Quote the intent back to the user in the confirmation. They said it; show that you heard it. Do not add encouragement ("great goal!") — that is exactly the sycophancy the project's manifesto rejects.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.