profile
Skill136 starsChanged 3 days agoMITView on GitHub
Captures and structures the user's profile from CV + questionnaire to make decisions on their behalf. Ensures quality matching.
30 other projects use a version of this.
Adopted by
Most recently taken up by
- DataDog/dd-trace-pytheir own version, first seen Nov 16, 2026
- rafaym1/scholarship-desktheir own version, first seen Nov 16, 2026
- baekenough/oh-my-customcodetheir own version, first seen Nov 14, 2026
- qbeka/jev-job-searchtheir own version, first seen Nov 12, 2026
- nockawa/NockMontheir own version, first seen Nov 8, 2026
- cismontane-harris2642/signal-prospecting-kittheir own version, first seen Nov 1, 2026
Used with
Projects with it also use coverage, benchmark and 4 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
PydanticSkillcoverage
Fetch coverage diff from Codecov for the current branch or a specific PR. Shows uncovered lines, patch coverage, and overall coverage change.
Also in 3 projects that use this
dragonflydbSkillbenchmark
Benchmark Dragonfly (and compare against Valkey/Redis) on local or remote cloud instances, then produce a performance + memory report with charts and a raw-data appendix. Use this whenever the user runs /benchmark, or asks to benchmark / load-test / measure throughput, QPS, latency, or memory efficiency of Dragonfly or another Redis-compatible server, to compare Dragonfly vs Valkey or Redis, or to generate a benchmark report. Trigger it even when the user just describes the goal ("see how fast Dragonfly is on this box", "compare memory per key vs Valkey") without saying the word "benchmark".
Also in 3 projects that use this
AnthropicSkillstatus
Case status summary by audience — client-facing (plain language), internal (for the professor), or court-ready (formal caption format per local rules). Same facts, different framing and depth. Use when a student needs to update the client, brief the professor, or prepare a court status report.
Also in 3 projects that use this
Microsoft AzureSkilldocs
Also in 3 projects that use this
VercelSkillsetup
Set up Vercel CLI and project configuration. Use when the user says "set up Vercel", "configure Vercel", "link to Vercel", or "vercel init".
Also in 3 projects that use this
GoogleSkilltdd
Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.
Also in 3 projects that use this
Other versions30 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 30
- ccusage/ccusage★ 19ksame name
- ryoppippi/ccusage★ 19ksame name
- FalkorDB/FalkorDB★ 11ksame name
- pytorch/executorch★ 5.2ksame name
- verl-project/verl-omni★ 1.5ksame name
- DataDog/dd-trace-py★ 655same name
- qbeka/jev-job-search★ 368same name
- nockawa/Typhon★ 253same name
- the-void-ia/void-box★ 88same name
- DataWithBaraa/claude-code-data-engineering-skills★ 84same name
20 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository15 other files it gives its agents.
- AGENTS.mdAGENTS.md
- acp-connectorSkill
- applySkill
- browser-automationSkill
- dailySkill
- dashboardSkill
- dbSkill
- memorySkill
- newsSkill
- onboardingSkill
- polishSkill
- radarSkill
- referralsSkill
- strategySkill
- targetsSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
13 rules. 0 are common, 9 are distinctive, and 4 are very specific to this project.
0
Common
Most of the 24 edited versions we read keep these.
9
Distinctive
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 32
Ask user for CV (URL or PDF).
- Distinctive: line 110
If clear → skip the question, save the inferred value with confidence note
- Distinctive: line 111
If unclear or missing → ask the question
- Distinctive: line 112
If the topic only applies to certain career stages (management questions) → skip if not applicable
- Distinctive: line 122
Ask the user directly. These are not inferable from a CV.
- Distinctive: line 154
Ask situation questions adapted to the user's career stage:
- Very specific: line 155
If career_stage is junior/mid: "Are you open to roles above your current level, or only same-level matches?" instead of "IC or Manager?"
- Very specific: line 156
If career_stage is senior+: "Would you accept IC roles or only Manager?" (only if has_management = true)
- Distinctive: line 165
After the strategy is defined, the agent suggests aligning the CV and LinkedIn profile to the job target to maximize match chances.
- Very specific: line 176
If the user agrees, trigger the polish skill which:
- Very specific: line 188
If the user says yes → run polish flow.
- Distinctive: line 189
If the user says no or later → skip, but remind them once at the end of onboarding.
- Distinctive: line 238
If user already has a profile in DB, validate changes before overwriting
“Common” is measured: kept by at least 6 in 10 of the 24 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Profile
Pre-flight
- Load active preferences (see
memoryskill):node scripts/db.js "SELECT category, key, value, confidence, source FROM preferences WHERE user_id = <user_id> AND status = 'active' ORDER BY category, key" - Load existing profile if present:
node scripts/db.js "SELECT data->'profile' AS profile, data->'job_preferences' AS job_preferences, data->'style_profile' AS style_profile FROM users WHERE id = <user_id>" - If profile exists, validate changes before overwriting
Flow overview
Step 1: CV analysis → extract experience, sector, profile, inferred seniority
Step 2: Gap questionnaire → only what the CV doesn't clarify (adaptive to inferred profile)
Step 3: Current situation + expectations → employment status, urgency, salary, work mode, availability
Step 4: Strategy → targets, sources, aggressiveness (informed by everything above)
Step 5: Polish suggestion → align CV and LinkedIn profile to the job target to maximize matchesStep 1: CV analysis
Ask user for CV (URL or PDF). Extract:
- Full name and title/profession
- Professional summary (elevator pitch)
- Work experience (company, role, period, achievements, team size, reporting line if applicable)
- Core competencies and tools (tech stack, software, methodologies, equipment, whatever is relevant to the field)
- Soft skills (leadership, communication, etc.)
- Certifications and courses
- Education (degrees, institutions)
- Languages and proficiency level
- Quantifiable achievements (metrics, impact)
- Notable projects or relevant work samples
Step 1b: Inferred profile
From the CV data, infer:
| Signal from CV | Inferred field | Used for |
| Years of experience, previous roles | career_stage (intern, junior, mid, senior, staff, principal, director+) | Which questionnaire blocks to show, seniority filtering |
| Team size managed, titles with Lead/Manager/Head | has_management (bool) | Whether to show management-related questions |
| Core competencies and tools | core_skills | Step 2 gap questions on skill preferences |
| Industries of previous employers | industry_history | Step 2 gap questions on industry preferences |
| Company sizes (startup vs corporate) | company_size_history | Step 2 gap questions on company size |
| Sector or functional area | sector | Step 2 gap questions, platform tiering |
| Languages and publications/conferences | visibility_level | Outreach tone, referral strategy |
Save to users.data.profile as JSONB, including the inferred fields:
node scripts/db.js "UPDATE users SET data = jsonb_set(data, '{profile}', '<json>'::jsonb) WHERE id = <user_id>" --writeThe inferred profile is preliminary. Step 2 confirms or corrects it.
Step 2: Gap questionnaire
Not a fixed set of blocks. The agent generates questions based on what the CV already answered clearly. Only ask about gaps.
Each answer must have a weight: Must (non-negotiable), Strong (strong preference), Nice (would be a plus).
Core questions (always ask if the CV doesn't clarify)
Role and career direction
- Role type (IC, manager, mixed, specialist, leadership) — adapt options to
career_stageandhas_management - Career stage confirmation (the agent proposes the inferred value, user confirms or corrects)
- Growth direction (stay in current path, wants to pivot, wants to move to leadership, unsure)
- Mentorship expectations (wants mentorship, wants autonomy, indifferent) — only if
career_stageis junior/mid
Management and autonomy (only if has_management = true or user expressed leadership aspirations)
- Expected reporting line (CEO, CTO/COO, VP, other director) — adapt to sector
- Expected hiring/firing authority
- Budget authority (own budget decisions)
- Org scope (single team, multiple squads, department, company-wide)
Work mode and geography
- Work mode (remote, hybrid, on-site)
- Current location and willingness to relocate
- Accepted timezones (Americas, Europe, Asia, global)
- Contract type (employee, contractor, freelancer)
Company type and sector
- Company size (startup, scale-up, mid-size, corporate)
- Company stage (early-stage, growth, established, public) — only if relevant to the sector
- Preferred sectors or industries (if
industry_historyshows concentration, confirm rather than ask open-ended) - Sectors to avoid (with nuance: absolute or accepts partial exposure?)
- Product/service type (own product, internal platform, consulting, services) — adapt to field
Sector-specific focus
- Area of specialization within the field (the agent proposes based on CV, user confirms or refines)
- Tools/methodologies the user wants to keep using vs open to learn
- Type of impact desired in first 6 months (adapt to role type)
Deal-breakers
- Graduated deal-breakers (the agent presents an empty list, the user adds their own). Never pre-load deal-breakers like "junior" or any role level. Each user defines their own.
How the agent decides what to ask
- For each topic above, check if the CV already provides a clear answer
- If clear → skip the question, save the inferred value with confidence note
- If unclear or missing → ask the question
- If the topic only applies to certain career stages (management questions) → skip if not applicable
- Present questions in blocks of 4, multi-select where applicable
Save to users.data.job_preferences as JSONB with weights:
node scripts/db.js "UPDATE users SET data = jsonb_set(data, '{job_preferences}', '<json>'::jsonb) WHERE id = <user_id>" --writeStep 3: Current situation and expectations
Ask the user directly. These are not inferable from a CV.
Employment status and urgency
- Current situation (employed active change, employed passive, unemployed, about to be unemployed, first job)
- How urgent is the search? (no urgency, in the coming months, soon, desperate)
- Availability to start (immediate, 2 weeks, 1 month, 3 months)
Compensation
- Salary range (min, expected, currency)
- Equity/participation expectations (if applicable to sector and career stage)
- Important benefits (health, education budget, equipment, etc. — adapt to sector)
- Flexibility on must-haves (e.g: 100% remote absolute or accepts 1 quarterly trip)
Availability for interviews
- Preferred time slot for interviews (e.g: "13:00 to 16:00 AR")
- Fixed blocked days/times (e.g: "Tuesday 14:00 to 15:00, English class")
- Timezone (default:
<Your IANA Timezone>)
Save to users.data:
node scripts/db.js "UPDATE users SET data = jsonb_set(data, '{availability}', '{\"preferred_hours\":\"<start>-<end> AR\",\"timezone\":\"<tz>\",\"blocked\":{\"<day>\":\"<start>-<end> (<reason>)\"}}') WHERE id = <user_id>" --writeThe news flow uses this to filter available slots from scheduling links without asking the user each time.
Step 4: Strategy
Now that the agent has the full profile (CV + gaps + situation), it can propose a strategy informed by everything above.
See the strategy skill for the full flow. Summary:
- Show the 4 levels (passive, selective, active, aggressive)
- Ask situation questions adapted to the user's career stage:
- If
career_stageis junior/mid: "Are you open to roles above your current level, or only same-level matches?" instead of "IC or Manager?" - If
career_stageis senior+: "Would you accept IC roles or only Manager?" (only ifhas_management = true)
- If
- Propose a level based on answers
- Allow customization of individual parameters
- Save to DB (
preferences.workflow.strategy_level+users.data.strategy)
The strategy's relax_must_haves resolves dynamically from the user's Must-weighted preferences (see AGENTS.md "Strategy levels"). No hardcoded keys.
Step 5: Polish suggestion
After the strategy is defined, the agent suggests aligning the CV and LinkedIn profile to the job target to maximize match chances.
What to do
- Compare the user's current CV and LinkedIn profile (from
users.data.profileandusers.data.linkedin_profile) against the defined job target (role, seniority, sector, must-haves) - Identify gaps:
- Keywords missing that recruiters search for
- Titles or descriptions that don't align with the target role
- Skills underrepresented relative to what the target market demands
- LinkedIn headline/summary that doesn't position the user for the target
- Present the analysis to the user
- If the user agrees, trigger the
polishskill which:- Audits the CV and LinkedIn profile section by section
- Drafts improvements aligned with the target
- Applies changes with per-section approval
- Exports a polished CV to PDF
How to present it
One short message, conversational (Gold Rule 7 applies):
"Ahora que sabemos que buscas
<role>en<sector>, tu CV y perfil de LinkedIn tienen algunas cosas que se pueden alinear mejor para que te encuentren más fácil. Querés que haga una revisión y te proponga cambios?"
If the user says yes → run polish flow.
If the user says no or later → skip, but remind them once at the end of onboarding.
Phase: Voice and style
This runs as part of Step 1 or as a separate phase after Step 5. It's independent of the career stage changes.
Automatic inference
- Open headless browser with persistent profile
- Read last 20-50 sent messages on LinkedIn (filter "You:")
- Read relevant sent emails in Gmail (to recruiters, HR, companies)
- Infer: tone, default language, writing characteristics, average length
- Extract 3-5 representative samples (1-2 recruiter, 2-3 personal)
Confirmation with options
Ask the user:
- Tone (formal, casual-professional, casual, direct/no-nonsense)
- Default language (Spanish, English, depends on context)
- Preferred length (short 1-3 lines, medium 4-6, long 7+)
- Preferred greeting (Hi [name], Dear, no greeting, other)
- Preferred closing (Regards, Cheers, no closing, other)
- Use bullet lists in messages? (yes, no)
- Emojis in professional messages? (yes, no, only in personal)
Validation
- Save inference + preferences to
users.data.style_profileas JSONB - Show 3 messages drafted with the style to the user for validation
- If user corrects → update style_profile
Phase: Platforms (output, not input)
- Consult
PLATFORMS.md - Cross-reference user profile vs role types/industries/seniority of each platform
- Assign Tier 1/2/3 to platforms based on fit
- Save to
users.data.platformsas JSONB - Don't ask the user. This is the output of analysis
Rules
- CV is source of truth. Questionnaire covers only what the CV doesn't say
- CV-first inference drives the questionnaire. The agent analyzes the CV, infers a preliminary profile, and generates only the gap questions that matter for that specific profile
- Career stage determines which questions appear. Management questions only for senior+ or users with management experience. Mentorship questions only for junior/mid. The repo never assumes a career stage
- No hardcoded deal-breakers. The deal-breakers list starts empty. Each user adds their own. Never pre-load role levels (junior, senior, etc.) as system deal-breakers
- Sector-agnostic. The profile flow works for any field (software, design, marketing, finance, operations, etc.). Never assume dev-specific concepts (tech stack, RAG, agents, IC) as universal. Adapt terminology to the user's field
- All DB access via
scripts/db.js(seedbskill). Read-only by default,--writefor saves - Persist everything to
users.dataas JSONB viajsonb_set:node scripts/db.js "UPDATE users SET data = jsonb_set(data, '{profile}', '<json>'::jsonb) WHERE id = <user_id>" --write node scripts/db.js "SELECT data->'profile' AS profile FROM users WHERE id = <user_id>" - If user already has a profile in DB, validate changes before overwriting
- Profile is updated when user changes CV or answers new questions
- Questions in blocks of 4, multi-select where applicable
- Each preference with weight: Must / Strong / Nice
- Platforms = output of analysis, never user input
- Single user (repo owner)
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
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.

