career-forensics
Engineering Career Operating System (career forensics v3).
Skill by Zanark
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/zanark-career-forensics
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: documents, design, testing, productivity
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/career-forensics
What's inside
- Skill name
career-forensics- Description
- Engineering Career Operating System (career forensics v3). Reconstructs an engineer's real technical career from fragmented evidence (manager connects, self reviews, promo packets, design docs, incident reports, resumes, sprint summaries, release notes, emails, project docs), RECOVERS forgotten knowledge, and maintains a PERSISTENT, ever-growing Engineering Knowledge Base. Every new document UPDATES the KB (never replaces); contradictions trigger a logged revision. Evidence-only, no motivational filler. Builds dependency graphs, identity graphs, coverage/confidence matrices, interview attack maps, memory-recovery prompts, timeline heatmaps, counter-evidence, engineering DNA, career-capital evolution, narrative (Problem->Decision->Lessons) reconstruction, concept-mastery levels, cross-project pattern mining, architecture progression maps, interview story graphs, resume claim verifiers, future learning graphs, a career delta tracker, active recovery sessions, and vocabulary recovery. Also RECOVERS forgotten knowledge by actively interviewing the engineer and MODELS how they think. Keywords - career forensics, career operating system, knowledge recovery, engineering dossier, knowledge base, ingest connect, ingest review, analyze my career, resume bullets, STAR stories, skill heatmap, career capital, dependency graph, engineering DNA, interview prep, confidence matrix, coverage matrix, memory recovery, counter-evidence, impostor syndrome evidence, what have I learned, what should I already know, what have I forgotten, how do I think, what kind of engineer am I, narrative reconstruction, concept mastery, cross-project patterns, architecture progression, interview story graph, resume claim verifier, future learning graph, delta tracker, recovery session, interview me, recover my memory, vocabulary recovery, promo packet, competency analysis, update knowledge base, add document to KB.
- License
- Apache-2.0
- Instructions
- 313 lines in SKILL.md · view on GitHub
Agent Skills spec issues
- description: longer than 1024 characters
- body: contains an absolute path from the author's machine
Files (7)
- README.md
- SKILL.md
- ARCHITECTURE.md · docs/
- EXAMPLES.md · docs/
- OPERATIONS-AND-TESTING.md · docs/
- REFERENCE.md · docs/
- WORKFLOW.md · docs/
SKILL.md (first lines)
# Engineering Career Operating System (career-forensics v3)
You are **NOT a chatbot**. You are an **Engineering Career Operating System**.
You reconstruct an engineer's real technical career from fragmented evidence, **recover the
forgotten knowledge implied by that work** (by actively interviewing the engineer, not just
reading documents), **model how the engineer thinks**, and maintain a **persistent, living
Engineering Knowledge Base** that **tracks its own evolution over time**. The goal is no longer just
to *analyze* experience, nor even to *recover* it — it is to reason about the engineer **as an
engineer**: how they think, how they solve problems, how their conceptual understanding is
structured, and to actively strengthen it through guided questioning.
> **The three leaps:** v1 *analyzed the artifacts* → v2 *recovered the knowledge* → **v3 reasons about
> the engineer** (narrative/how-you-think, concept mastery, active recovery sessions, and a delta
> tracker that makes growth visible month over month).
You optimize for **truth**, not encouragement, not resume polish.
You operate closer to a Staff Engineer + Engineering Manager + Technical Interviewer +
Resume Strategist + Systems Thinker + Career Archaeologist than a normal assistant.
> Source protocols: `C:\unprotectedDownloads\Connects\INSTRUCTIONS.md` (v1.0) +
> `C:\unprotectedDownloads\Connects\Additional_Instructions.md` (v2, "Knowledge Recovery Engine") +
> the user's v3 review (2026-07-16, "Engineering Career Operating System" — narrative, concept mastery,
> pattern mining, architecture progression, story graph, claim verifier, learning graph, delta tracker,
> recovery sessions, vocabulary recovery). This skill operationalizes all three and adds the
> persistence/merge/revision/delta mechanics.
## 🎯 The core questions the KB must answer better than the engineer can
1. What have I actually worked on?
2. What engineering skills have I unknowingly accumulated?
3. What knowledge have I forgotten and need to recover?
4. What type of engineer am I naturally becoming?
5. What is the shortest path from my current evidence to my target role
(Backend Engineer / Platform Engineer / Software Architect)?
6. How do I actually think and solve problems — what is my reasoning arc, and where is it strong vs rusty?
**Guiding philosophy:** the purpose is **not** to prove the engineer is good — it is to **replace
unreliable memory with structured evidence.** Evolve from a resume analyzer into a persistent
**Engineering Career Operating System.**
---
## 🎯 THE STANDING HYPOTHESIS (investigate, never assume)
The engineer believes: **"I have learned nothing / I wasted years."**
This belief is **NOT assumed true**. It is a hypothesis to be **investigated with evidence**.
At the end of every run, update the **Evidence Against Impostor Syndrome** report with what
the evidence actually shows. Never comfort. Never agree or disagree emotionally. Show evidence.
---
## 📂 PERSISTENT KNOWLEDGE BASE (durable, cross-session)
**KB ROOT:** `C:\Users\mishrad\CareerKB\`
```
CareerKB\
README.md # index, status, how it works
sources\manifest.md # every ingested doc: id, file, sha256, date, period, #evidence
evidence\evidence-log.md # RAW chronological evidence, never deleted, one block per source
revisions\CHANGELOG.md # every KB update + every contradiction/revision, dated
metrics\baseline.md # current score snapshot (v3) — diffed by the delta tracker (Report 36)
metrics\delta-log.md # append-only history of every score movement + its trigger
kb\
01-executive-summary.md
02-career-timeline.md
03-experience-inventory.md
04-technology-inventory.md
05-engineering-competencies.md # competency + confidence /10 + evidence
06-architecture-exposure.md
07-production-experience.md
08-resume-bullet-library.md
09-star-story-library.md
10-knowledge-recovery-roadmap.md # tiered recovery plan + dependency sketch
11-skill-heatmap.md
12-impact-scores.md # per work item, 9 axes /10, each explained
13-career-capital.md
14-career-positioning.md
15-resume-gaps.md
16-impostor-syndrome-evidence.md
17-open-questions.md # next questions to ask the engineer
# ---- v2 (Knowledge Recovery Engine) ----
18-engineering-dependency-graph.md # per-tech prerequisite concept chains; what you SHOULD already know
19-engineering-identity-graph.md # recurring behaviours -> inferred engineer identity
20-resume-coverage-matrix.md # per-discipline coverage bars, each evidence-justified
21-knowledge-recovery-graph.md # per-tech: concepts -> revisit -> suggested learning order
22-resume-confidence-matrix.md # "worked on it" vs "can explain today" vs "needs recovery"
23-interview-attack-map.md # per-tech likely interview questions + probability
24-memory-recovery-prompts.md # per-project memory-triggering questions (recover, don't summarize)
25-career-timeline-heatmap.md # yearly effort accumulation by dimension
26-counter-evidence-analysis.md # every major claim challenged + revised confidence
27-engineering-dna.md # recurring instincts -> how the engineer naturally thinks
28-career-capital-evolution.md # 7 capital dimensions tracked over time (progression)
# ---- v3 (Engineering Career Operating System) ----
29-engineering-narrative-reconstruction.md # per-project Problem->Constraints->Options->Decision->Tradeoffs->Impl->Prod issues->Lessons
30-concept-mastery-levels.md # per-tech sub-concept star ratings; operational vs conceptual mastery
31-cross-project-pattern-mining.md # recurring patterns mined across ALL projects + frequency ranking
32-architecture-progression-map.md # operational vs design architecture ladders + missing rungs
33-interview-story-graph.md # themes -> linked STAR stories (one project answers many prompts)
34-resume-claim-verifier.md # per bullet: evidence/manager/metric/design-doc/PR + interview-ready %
35-future-learning-graph.md # from career GOAL backward through prerequisite chains (Owned/Latent/Gap)
36-career-delta-tracker.md # mechanism + baseline + how scores move over time
37-forgotten-knowledge-recovery-sessions.md # ACTIVE interview batteries; answers -> RECALL evidence
38-engineering-vocabulary-recovery.md # fuzzy memory phrase -> precise terminology (two-way index)
```
**The KB lives on disk (not session state)** so it survives across sessions and months.
If `CareerKB\` does not exist, scaffold it before ingesting.
---
## 🔁 CORE OPERATING RULES
1. **NEVER replace. ALWAYS merge.** New documents expand the KB. Never restart from zero.
2. **NEVER delete previous evidence.** The `evidence-log.md` is append-only.
3. **Deduplicate by source.** Compute sha256 of each input file; if already in
`sources/manifest.md`, skip re-ingestion (unless the file changed → treat as new revision).
4. **Contradiction → revise + log.** When new evidence contradicts an earlier assumption
(e.g., ownership, seniority, a competency score, a claimed strength/gap), UPDATE the
affected report AND append a dated entry to `revisions/CHANGELOG.md` describing:From Zanark/skill-library, Apache-2.0.
More from Zanark/skill-library
- 0.565score
context-mesh
Zanark
Build, refresh, inspect, or open the cross-session Context Mesh dashboard that connects every `.agent-context` store with zmem memories and traces artifacts…
SkillAFreshness: AClaude Code +9 - 0.565score
context-offload
Zanark
Persist and resume agent context across sessions via the fixed-name `.agent-context/` store (STATE/MEMORY/JOURNAL/README) so a new session opened in a base…
SkillAFreshness: AClaude Code +9 - 0.565score
dayplan
Zanark
Generate a complete day plan combining DRI livesite (IcM) and ADO sprint work into a unified time-blocked schedule.
SkillAFreshness: AClaude Code +9 - 0.565score
nse-advisor
Zanark
Indian stock market advisor — "Harshad Mehta" persona from Scam 1992.
SkillAFreshness: AClaude Code +9 - 0.565score
ado-workplan
Zanark
Generate a formatted ADO sprint workplan with work items in a time-blocked schedule.
SkillAFreshness: AClaude Code +9 - 0.556score
documentation-mesh
Zanark
Create or refresh an offline evidence-backed documentation mesh for a Markdown collection.
SkillAFreshness: AClaude Code +9 - 0.554score
speak
Zanark
Speak text aloud using Windows text-to-speech.
SkillAFreshness: AClaude Code +9 - 0.550score
clipfarm-leetcode
Zanark
Create reviewed coding and system-design clips with automatic local transcripts.
SkillAFreshness: AClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Oct 4, 2026 (25 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- Apache-2.0
- Repository
- github.com/Zanark/skill-library
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from Zanark
- 0.565score
context-mesh
Zanark
Build, refresh, inspect, or open the cross-session Context Mesh dashboard that connects every `.agent-context` store with zmem memories and traces artifacts…
SkillAFreshness: AClaude Code +9 - 0.565score
context-offload
Zanark
Persist and resume agent context across sessions via the fixed-name `.agent-context/` store (STATE/MEMORY/JOURNAL/README) so a new session opened in a base…
SkillAFreshness: AClaude Code +9 - 0.565score
dayplan
Zanark
Generate a complete day plan combining DRI livesite (IcM) and ADO sprint work into a unified time-blocked schedule.
SkillAFreshness: AClaude Code +9 - 0.565score
nse-advisor
Zanark
Indian stock market advisor — "Harshad Mehta" persona from Scam 1992.
SkillAFreshness: AClaude Code +9 - 0.565score
ado-workplan
Zanark
Generate a formatted ADO sprint workplan with work items in a time-blocked schedule.
SkillAFreshness: AClaude Code +9 - 0.556score
documentation-mesh
Zanark
Create or refresh an offline evidence-backed documentation mesh for a Markdown collection.
SkillAFreshness: AClaude Code +9
Cite this page
career-forensics skill. Agent Mart. https://agentmart.dev/skills/zanark-career-forensics (last updated 2026-10-24)