Add jobstead skill#2285
Closed
adibas03 wants to merge 1 commit into
Closed
Conversation
Persistent, session-spanning job-search companion: fit-checks a posting against a stored applicant profile, flags scam patterns, and tailors a resume/cover letter grounded in the person's real background rather than generic one-shot advice. Profile, application tracker, and accumulated lessons persist across sessions in references/. Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01Q5CnBdJyAAY3g11qj7i8ha
Contributor
🔒 PR Risk Scan ResultsScanned 6 changed file(s).
|
Contributor
🔍 Vally Lint Results✅ All checks passed
Summary
Full linter output |
Contributor
|
This doesn't really fit with awesome copilot at the moment. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Persistent, session-spanning job-search companion: fit-checks a posting against a stored applicant profile, flags scam patterns, and tailors a resume/cover letter grounded in the person's real background rather than generic one-shot advice. Profile, application tracker, and accumulated lessons persist across sessions in references/.
Claude-Session: https://claude.ai/code/session_01Q5CnBdJyAAY3g11qj7i8ha
Pull Request Checklist
skills/jobstead/).namein frontmatter, lowercase-hyphenated).npm startand verified that the generated docs (docs/README.skills.md) are up to date.mainbranch for this pull request.Description
Adds
jobstead, a persistent, session-spanning job-search companion skill. Unlike a one-shot task helper, it grounds every fit-check and story in accumulated state that persists across sessions:references/profile.md— the applicant's background, work authorization, target roles, and constraints (ships blank; the skill asks for a resume on first use and fills it in)references/tracker.md— an append-only application tracker (applied/rejected/offered/deferred, etc.)references/lessons.md— accumulated, identity-free patterns learned across the search (seeded with a starter playbook covering visa/sponsorship signals, salary-floor plausibility checks, ATS-tool cost/benefit, and cover-letter framing)references/log.md— a per-session activity auditCore capabilities: fit-checking a posting against the stored profile (not generic advice), catching scam patterns in postings/recruiter messages, and building a story-grounded, ATS-optimized resume/cover letter tied to a specific role.
Relationship to
technical-job-searchThis repo already has a
technical-job-searchskill, which is explicitly one-shot/stateless per task (JD analysis, CV tailoring, cover letter, offer evaluation, follow-up email — no persistence across sessions).jobsteadis a different, complementary tool: everything it does is grounded in a profile/tracker/lessons state that accumulates over the whole job search rather than being re-derived each time, and it explicitly covers scam-detection as a first-class use case.Type of Contribution
Additional Notes
The trigger description was validated with an extensive eval methodology (sequential nested-session testing across a 60-query set of should-trigger/should-not-trigger phrasings, with full transcript capture rather than pass/fail booleans): 86% recall, ~100% precision on fair, well-formed queries. Happy to share the eval harness/results if useful for review.
By submitting this pull request, I confirm that my contribution abides by the Code of Conduct and will be licensed under the MIT License.