Install
$ agentstack add skill-neuroanalytics-data-science-harness-obligations ✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README — it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming — see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps — measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Skill: obligations
Maintain the project's obligations — the commitments it must discharge (a frozen confirmatory comparison that must be run as specified, a Data Management Plan deliverable, an ethics amendment, a funder report due by a date). This is the "to-do list of promises" of the Manage & Comply lane, tracked in the ledger obligations[] so the administrative record is provenanced alongside the science. You own the tracking judgment; you delegate the save to the datalad doer.
When to use
- The user wants to see what is outstanding/overdue, add a new commitment, or mark one met/waived.
- Do NOT use to create a pre-registration (that is
govern/preregister, which adds the
obligation) or to run the confirmatory analysis (analyze/run-comparison).
Steps
- Read the current obligations — from
project.yamlobligations[]. Present a concise
surface: what is pending (highlight anything with a due date that is near or past), what is met, what is waived. Group by kind (preregistration, confirmatory-comparison, dmp, ethics, funder-report).
- Act on the request:
- Add — append
{ id, kind, description, due?, status: pending, ref? }(uniqueid). - Resolve — flip an entry's
statustomet(the commitment was fulfilled — e.g. the
confirmatory comparison was completed as specified) or waived (dropped, with the reason in the log). Never delete an obligation; the record of the commitment stays.
- Update — add a
refonce a pending registration is assigned an id/URL, or adjust adue.
- Log it — append
{ ts, op: obligations, stage: govern, note: " ", branch: }. - Save — delegate to the datalad doer: "save:
datalad save -m 'obligations: '." - Report — the updated surface, and the next thing due (by date, then by kind). If a
confirmatory-comparison obligation is still pending, point to analyze/run-comparison; if a pre-registration lacks a ref, point to govern/preregister.
Constraints
- Obligations are never deleted — resolution is a forward
statuschange (pending→met/waived)
with the reason captured in the log. This preserves the compliance trail.
- Keep
obligations[]ids unique; add/resolve perdocs/project-ledger.md. - Do not silently mark a confirmatory obligation
met— confirm the comparison was actually run as
specified (any deviation from the frozen spec is reportable, not a quiet pass).
- Keep
log:append-only and the ledger schema-valid; delegate the save to the datalad doer.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: neuroanalytics
- Source: neuroanalytics/data-science-harness
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet — be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.