Agent skill

Openfasttrace Reverse Specs

by itsallcode in itsallcode/openfasttrace

Reverse-engineer missing or incomplete OpenFastTrace system requirements and arc42-style design documentation from a project's user guide, existing documentation, tests, and code.

GPL-3.0Auto-check passedSecurity

Install Openfasttrace Reverse Specs

skills CLI
$ npx skills add itsallcode/openfasttrace --skill openfasttrace-reverse-specs -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install itsallcode/openfasttrace openfasttrace-reverse-specs --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/itsallcode/openfasttrace.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/openfasttrace-reverse-specs .claude/skills/openfasttrace-reverse-specs && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
openfasttrace-reverse-specs
GitHub stars
198
Token cost
~2.9k tokens
SKILL.md length
1,479 words
Files
17 (incl. references, assets)
Skills in repo
3
Repo updated
First seen
Licence
GPL-3.0

At a glance

Reverse-engineer missing or incomplete OpenFastTrace system requirements and arc42-style design documentation from a project's user guide, existing documentation, tests, and code.

  • Works in 3 steps: Draft System Requirements → Draft Design → Add Coverage Markers in Implementation…
  • The agent must draft
  • SKILL.md covers Operating Rules, Evidence Order, Preconditions and Trace Commands, plus 6 more sections
  • Calls java

What it does

Openfasttrace Reverse Specs is an agent skill from itsallcode/openfasttrace. Reverse-engineer missing or incomplete OpenFastTrace system requirements and arc42-style design documentation from a project's user guide, existing documentation, tests, and code. Use when the agent must draft or repair doc/systemrequirements.md, doc/design.md, and doc/design/ chapters; infer features, requirements, scenarios, and design items; align design coverage with requirements; and report contradictions or open issues found during reverse engineering.

Its SKILL.md is about 2.9k tokens, which your agent loads only when the skill is triggered. The skill folder holds 20 other files, including reference files and assets (for example `agents/openai.yaml`, `assets/design/architecture_decisions.md` and `assets/design/building_block_view.md`).

It sits in Security, covering Technical writing, Design tokens and Reverse engineering and malware. It works with Java. The repository describes itself as: Open source requirement tracing suite. The licence is GPL-3.0.

When your agent uses it

  • The agent must draft
  • Repair doc/systemrequirements.md
  • Doc/design/ chapters
  • Align design coverage with requirements

Example prompts

  • “/openfasttrace-reverse-specs”

Workflow steps

3 steps, taken from the step headings in SKILL.md.

  1. Draft System Requirements
  2. Draft Design
  3. Add Coverage Markers in Implementation and Tests

What it can do on your machine

Read from SKILL.md and the folder at commit be8537e. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • java

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Openfasttrace Reverse Specs loads about 2.9k tokens when it runs, and up to ~3.5k if it reads all its reference files. Until then it costs about 124 tokens; SKILL.md has 1,479 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~124
When it runs · the whole SKILL.md, loaded when a task matches
~2.9k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~3.5k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from itsallcode/openfasttrace at commit be8537e, republished under its GPL-3.0 licence (© itsallcode). 1,479 words, ~2,914 tokens.

Download SKILL.mdSave it as .claude/skills/openfasttrace-reverse-specs/SKILL.md (or your agent's skills folder). This skill also uses 16 other files; get the full folder from GitHub.
name
openfasttrace-reverse-specs
description
Reverse-engineer missing or incomplete OpenFastTrace system requirements and arc42-style design documentation from a project's user guide, existing documentation, tests, and code. Use when the agent must draft or repair `doc/system_requirements.md`, `doc/design.md`, and `doc/design/` chapters; infer features, requirements, scenarios, and design items; align design coverage with requirements; and report contradictions or open issues found during reverse engineering.

OpenFastTrace Reverse Specs

Reverse-engineer OFT specifications from existing project artifacts. Treat user-facing documentation as the primary source for intent and source code as evidence for implemented behavior and design.

Use the bundled templates when creating new documentation:

  • assets/system_requirements_template.md
  • assets/design_index_template.md
  • assets/design/*.md

For OFT syntax and trace behavior, use the openfasttrace-skill if it is available in the session.

Operating Rules

  • DO NOT change any code. The only exception are coverage marker comments in Step 3.
  • DO NOT change the documents from which you reverse engineer specification (user guide, README).
  • Preserve existing project specification documents unless the user explicitly asks to replace them. If a target file exists, patch it carefully or create a clearly named draft beside it.
  • Keep a source inventory while working. For every inferred requirement or design item, know whether it came from user guide, README, tests, source code, configuration, build files, issue text, or runtime behavior.
  • Prefer explicit evidence over speculation. Mark weakly supported conclusions in Open Issues.
  • Do not hide contradictions. Summarize them in the final answer and record them in the generated document's Open Issues section.
  • Use stable OFT item IDs in lower-kebab-case unless the project already uses another convention.
  • Use revision ~1 for newly inferred items unless replacing an existing item with a semantically incompatible version.
  • Keep generated text in draft status when confidence is incomplete.

Evidence Order

Inspect artifacts in this order:

  1. User guide, manual, tutorial, screenshots, examples, CLI help, public API docs, README usage sections, and release notes.
  2. Existing requirements, design, architecture decision records, issue plans, and trace files.
  3. Tests, examples, fixtures, and golden files, especially acceptance, integration, CLI, API, and UI tests.
  4. Public entry points in code: commands, controllers, UI actions, service APIs, plugin extension points, exported classes, configuration keys, and error messages.
  5. Internal code structure, dependency declarations, build scripts, packaging, deployment manifests, and runtime integration points.

The user guide is closest to product intent. If code and user guide disagree, prefer neither silently. Record the contradiction and ask the user to decide.

In absence of a user guide, tests are the next best option to extract product intent.

Preconditions

OpenFastTrace (OFT) should be available to be able to check the created specifications.

If it is not installed, download the JAR artifact from OFT's latest GitHub release (https://github.com/itsallcode/openfasttrace/releases) and use that for the trace.

Use oft in the commands below. If only the JAR is available, replace oft with java -jar /path/to/openfasttrace.jar.

Trace Commands

Run traces with the narrowest artifact scope for the current reverse-engineering phase. The -a switch is a whitelist of artifact types included in the scan.

System Requirements Only

Run this after Step 1 to check feature, requirement, and scenario links inside doc/system_requirements.md:

bash
oft trace -a feat,req,scn doc/system_requirements.md

Expected result: feature, requirement, and scenario links are complete. Downstream design coverage is intentionally out of scope in this trace, even when scenarios declare Needs: dsn.

System Requirements to Design

Run this after Step 2 to check coverage from system requirements down to design:

bash
oft trace -a feat,req,scn,constr,dsn doc/system_requirements.md doc/design.md doc/design

Expected result: scenarios are covered by design items or OFT forwarding notation. Implementation and test coverage are intentionally out of scope.

Full Project Coverage

Run this after Step 3 to check coverage from requirements through design, implementation, tests, and build markers:

bash
oft trace .

Expected result: the full requirement network is covered across all project files.

Decision Checkpoints

Use decision checkpoints to resolve open issues without derailing the reverse-engineering pass.

At the end of Step 1 and Step 2, summarize unresolved issues as a numbered list of concrete questions. For each question, include the conflicting or missing evidence, the decision needed, and a recommended default when the evidence supports one. If the user answers, update the document and remove resolved issues. If the user cannot answer or does not answer yet, keep the issue in Open Issues and keep the affected items in draft status.

Ask earlier only when a contradiction blocks writing a traceable draft.

Step 1: Draft System Requirements

Create or update doc/system_requirements.md using assets/system_requirements_template.md.

Build the document in this order:

  1. Write the introduction from the user guide and README, not from implementation structure.
  2. Define terms and user roles before feature details.
  3. Extract product-level feat items from user-visible capabilities. Extract features sparingly. The level of granularity should be what would be listed on a product flyer.
  4. Refine each feature into req items that describe user-visible needs and constraints.
  5. Add scn items as Given-When-Then acceptance scenarios.
  6. Add Needs and Covers links:
    • feat items need req.
    • req items cover feat and need scn.
    • scn items cover req and need dsn.
  7. Fill gaps from tests and code only after user-guide-derived intent is represented.
  8. Record uncertain inferences, missing intent, duplicate behavior, and contradictions in Open Issues.
  9. Run a Decision Checkpoint for open issues that affect system requirements.
  10. Run the System Requirements Only trace. Verify the trace is clean for the included artifact types.
  11. After drafting system requirements, stop for user review unless the user explicitly requested a complete requirements-and-design reverse-engineering pass in one turn. Ask the user to remove the draft marker from requirements they reviewed and approved.
  12. Report draft / total counts by artifact type and overall.
Show full SKILL.md (622 more words)Show less

Step 2: Draft Design

  1. Create or update doc/design.md and doc/design/ using:

    • assets/design_index_template.md
    • files under assets/design/
  2. Derive design from code and tests, then match it against doc/system_requirements.md.

    Use this design structure:

    • Introduction and goals in doc/design.md
    • Architecture constraints
    • Context and scope
    • Solution strategy
    • Building block view
    • Runtime view
    • Deployment view
    • Crosscutting concepts
    • Architecture decisions
    • Quality requirements
    • Risks and technical debt
    • Glossary
    • Open issues
  3. Add dsn items where design decisions or runtime behavior cover scn and constr items. Prefer one runtime dsn per scenario or constraint. If a design layer adds no information, use OFT forwarding notation rather than inventing redundant design text.

  4. If the documents contain information about intentional technical constraints, document them as constr in section Architecture constraints. Each constr item needs dsn coverage, and at least one dsn item must cover it. For example the project documentation states that only builds for x86 are supplied. Try not to infer constraints from the code, because they might not be intentional.

  5. Record contradictions under Open Issues, especially when:

    • a scenario from system requirements is not implemented by the observed design,
    • code implements behavior with no requirement,
    • design structure conflicts with user-facing intent,
    • tests assert behavior that differs from user-guide wording,
    • public configuration or API behavior is undocumented,
    • dependencies, persistence, network access, security, or deployment behavior contradict stated constraints.
  6. Run a Decision Checkpoint for open issues that affect design or design coverage.

  7. Run the System Requirements to Design trace. Verify the trace is clean for the included artifact types.

  8. After drafting the design, stop for user review unless the user explicitly requested a complete requirements-and-design reverse-engineering pass in one turn. Ask the user to remove the draft marker from requirements they reviewed and approved.

  9. Report draft / total counts by artifact type and overall.

Step 3: Add Coverage Markers in Implementation and Tests

Do not change code logic. In this step, only add OFT coverage markers as comments.

  1. For each dsn item, read its Needs field and add exactly the requested downstream marker types:

    • impl for implementation code
    • utest for unit tests
    • itest for integration tests
    • stest for system tests
    • bld for build configuration
  2. Use the OFT tag notation [<covering-type>-><covered-item-id>] in the host file's comment syntax. Examples:

    • Java, JavaScript, TypeScript, C, C++, C#, Go: // [impl->dsn~drop-database-objects~1]
    • Python, shell, YAML, TOML: # [utest->dsn~drop-database-objects~1]
    • SQL: -- [itest->dsn~drop-database-objects~1]
  3. Place each marker directly above the smallest stable code element that implements or verifies the design item: method, function, class, test method, test class, build block, or configuration entry. Prefer narrow locations over file-level markers.

  4. Do not add marker types that are not listed in the dsn item's Needs. If the observed project lacks the requested implementation or test evidence, do not invent coverage. Record the missing coverage in the design document's Open Issues section.

  5. Run the Full Project Coverage trace. Verify the trace is clean.

Reverse-Engineering Checklist

Use references/evidence_checklist.md when the project is large or unfamiliar.

For each feature area, collect:

  • intent source: guide, README, API docs, screenshots, examples, or release notes
  • behavioral evidence: tests, examples, CLI output, UI actions, endpoint definitions, logs, errors
  • design evidence: modules, packages, extension points, data flow, persistence, network calls, external systems
  • verification evidence: unit tests, integration tests, acceptance tests, build gates
  • confidence: high, medium, or low
  • open issues: contradictions, missing source, unclear terminology, scope questions

Output Discipline

When reporting back to the user:

  • summarize created or changed files,
  • list contradictions and open issues first if any exist,
  • state which evidence sources were used,
  • state whether the next step is user review or design extraction,
  • mention if OFT tracing was not run.

Do not claim the reverse-engineered specification is complete. It is a structured draft until the user resolves intent and contradiction questions.

© itsallcode, GPL-3.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 16 other files (references, assets) in .agents/skills/openfasttrace-reverse-specs of itsallcode/openfasttrace.

  • SKILL.md
  • agents/openai.yaml
  • assets/design/architecture_decisions.md
  • assets/design/building_block_view.md
  • assets/design/constraints.md
  • assets/design/context_and_scope.md
  • assets/design/crosscutting_concepts.md
  • assets/design/deployment_view.md
  • assets/design/glossary.md
  • assets/design/open_issues.md
  • assets/design/quality_requirements.md
  • assets/design/risks_and_technical_debt.md
  • assets/design/runtime_view.md
  • assets/design/solution_strategy.md
  • assets/design_index_template.md
  • assets/system_requirements_template.md
  • references/evidence_checklist.md

Open the folder on GitHubat commit be8537e

Compare with similar skills

Openfasttrace Reverse Specs next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Openfasttrace Reverse Specs compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Openfasttrace Reverse Specs this skillitsallcode/openfasttrace198—~2.9kAutomated safety check: PassGPL-3.0
Java Decompilerptn1411/skill219—~788Automated safety check: NotesNone
APK Static Analysisdslsdzc/rev-skills117—~2kAutomated safety check: PassApache-2.0
Mintlify Claude Docslilinji/ai-infra-odyssey127—~8.2kAutomated safety check: PassNone
805 Regulations Eu Cyber Resilience Actjabrena/plinth445—~3kAutomated safety check: PassApache-2.0
Security Advisory Rewriterobot-platform/obot1.1k—~1.3kAutomated safety check: PassMIT

Similar skills

  • Java Decompiler

    ptn1411/skill

    Decompile Java applications (JAR/WAR/APK/class) — extract archives, detect obfuscators, decompile bytecode to source, analyse license logic and secrets.

    219 GitHub stars~788 tokensUpdated 15 days ago
    SecurityAuto-check: notes
  • APK Static Analysis

    dslsdzc/rev-skills

    Guides static analysis of an Android APK with jadx and apktool: reading the manifest, Java code, resources and permissions, and recognizing hardening or obfuscation.

    117 GitHub stars~2k tokensUpdated 2 days ago
    SecurityAuto-check passed
  • Mintlify Claude Docs

    lilinji/ai-infra-odyssey

    Build, write, review, migrate, and maintain Mintlify documentation in a restrained Claude Docs-inspired style.

    127 GitHub stars~8.2k tokensUpdated 9 days ago
    Writing & ContentAuto-check passed
  • A skill your agent uses when reviewing, designing, or modifying Java enterprise products, services, libraries, agents, plugins, connected components, or platform modules that may qualify as products…

    445 GitHub stars~3k tokensUpdated today
    SecurityAuto-check passed
  • Security Advisory Rewriter

    obot-platform/obot

    Turns a verbose, reporter-submitted obot security advisory into a short, deployer-facing writeup covering impact, affected versions and mitigation.

    1.1k GitHub stars~1.3k tokensUpdated today
    SecurityAuto-check passed
  • Dt Obs Services

    Dynatrace/dynatrace-for-ai

    Service performance monitoring with RED metrics (Rate, Errors, Duration) and runtime-specific telemetry for Java, .NET, Node.js, Python, PHP, and Go.

    161 GitHub stars~3.3k tokensUpdated 6 days ago
    DevOps & CloudAuto-check passed

More from itsallcode/openfasttrace

  • Oft Spec Driven Development

    itsallcode/openfasttrace

    Support spec-driven development in projects that use OpenFastTrace, a system-requirements document in doc/systemrequirements.md, an arc42-style design in doc/design.md, and per-issue task plans in…

    198 GitHub stars~2k tokensUpdated today
    Auto-check passed
  • Openfasttrace

    itsallcode/openfasttrace

    Work with OpenFastTrace requirement tracing, including specification items, artifact IDs, coverage markers, Markdown and Gherkin syntax, and trace validation.

    198 GitHub stars~1.6k tokensUpdated today
    Auto-check passed

Works with

Questions about Openfasttrace Reverse Specs

What does Openfasttrace Reverse Specs do?

Reverse-engineer missing or incomplete OpenFastTrace system requirements and arc42-style design documentation from a project's user guide, existing documentation, tests, and code. Openfasttrace Reverse Specs is an agent skill from itsallcode/openfasttrace. Reverse-engineer missing or incomplete OpenFastTrace system requirements and arc42-style design documentation from a project's user guide, existing documentation, tests, and code.

When should I use Openfasttrace Reverse Specs?

Openfasttrace Reverse Specs fits situations like: the agent must draft; repair doc/systemrequirements.md; doc/design/ chapters; align design coverage with requirements.

How do I install Openfasttrace Reverse Specs in Claude Code?

Run `npx skills add itsallcode/openfasttrace --skill openfasttrace-reverse-specs -a claude-code`. Or copy the skill folder (.agents/skills/openfasttrace-reverse-specs in itsallcode/openfasttrace) into .claude/skills/openfasttrace-reverse-specs in your project. Claude Code loads it when a task matches its description.

How do I install Openfasttrace Reverse Specs in Codex?

Run `npx skills add itsallcode/openfasttrace --skill openfasttrace-reverse-specs -a codex`. Or copy the skill folder (.agents/skills/openfasttrace-reverse-specs in itsallcode/openfasttrace) into .agents/skills/openfasttrace-reverse-specs in your project. Codex loads it when a task matches its description.

Can I use Openfasttrace Reverse Specs in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add itsallcode/openfasttrace --skill openfasttrace-reverse-specs -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/openfasttrace-reverse-specs, .gemini/skills/openfasttrace-reverse-specs, .github/skills/openfasttrace-reverse-specs and .opencode/skills/openfasttrace-reverse-specs in your project.

What does Openfasttrace Reverse Specs need to run?

Going by SKILL.md and its folder, Openfasttrace Reverse Specs needs the command-line tools its instructions call (java).

Does Openfasttrace Reverse Specs access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Openfasttrace Reverse Specs safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Openfasttrace Reverse Specs use?

Openfasttrace Reverse Specs is published under the GPL-3.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Openfasttrace Reverse Specs use?

About 2.9k tokens (SKILL.md is roughly 12k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 594 tokens, read only when the agent opens those files.

What are the alternatives to Openfasttrace Reverse Specs?

Skills that share tags, products or a category with Openfasttrace Reverse Specs: Java Decompiler (ptn1411/skill, 219 stars), APK Static Analysis (dslsdzc/rev-skills, 117 stars), Mintlify Claude Docs (lilinji/ai-infra-odyssey, 127 stars) and 805 Regulations Eu Cyber Resilience Act (jabrena/plinth, 445 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Openfasttrace Reverse Specs?

itsallcode (a GitHub organization) maintains it in itsallcode/openfasttrace, which has 198 GitHub stars. The repository holds 3 skills in this directory. The repository was last updated on October 7, 2026.

Source: itsallcode/openfasttrace on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.