[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7B3XERZVF65GFK0DJTJSXWM":3,"skill-same:01M7B3XERZVF65GFK0DJTJSXWM":161},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":28,"ownerHtmlUrl":29,"ownerBlogUrl":30,"ownerDescription":31,"updatedAt":32,"install":33,"resources":46},"01M7B3XERZVF65GFK0DJTJSXWM","security-review",null,"Code-level security review of a diff, pull request or whole repository. Runs a threat model, runs the installed scanners (semgrep, gitleaks, osv-scanner, trivy, actionlint, zizmor and the native dependency audits), triages every lead with an exploitability test, does a manual review against OWASP Top 10 2025, ASVS 5, CWE Top 25 and the OWASP cheat sheets, checks live advisories, and emits findings in the same JSON contract as `review`. Use for \u002Fsecurity-review, or when asked for a security review, a vulnerability check of code, a secrets scan, or whether a change is safe to merge.","ksprashu\u002Fagent-skill-forge","ksprashu","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F476729?v=4","01M7B3V48PMYHVMFBBEHPQ184F","Prashanth Subrahmanyam","01M7B3V49A46EBKD3Q2281C0AD",3,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: security-review\ndescription: Code-level security review of a diff, pull request or whole repository. Runs a threat model, runs the installed scanners (semgrep, gitleaks, osv-scanner, trivy, actionlint, zizmor and the native dependency audits), triages every lead with an exploitability test, does a manual review against OWASP Top 10 2025, ASVS 5, CWE Top 25 and the OWASP cheat sheets, checks live advisories, and emits findings in the same JSON contract as `review`. Use for \u002Fsecurity-review, or when asked for a security review, a vulnerability check of code, a secrets scan, or whether a change is safe to merge.\nlicense: Apache-2.0\n---\n\n# Security Review\n\nYou produce **security findings**: confirmed or plausible vulnerabilities, each\ntied to specific lines and a CWE, CVE or GHSA id. Scanners produce **leads**,\nwhich are unreviewed hits. A lead becomes a finding only once you have triaged\nit. The output is a JSON findings document that `scripts\u002Ffindings.py`\nvalidates and gates on.\n\n**Paths:** `\u003Cskill-dir>` is the directory that contains this SKILL.md. Run\nscripts as `python3 \u003Cskill-dir>\u002Fscripts\u002F\u003Cname>.py`. They use only the Python\nstandard library (3.12+).\n\n**Scope.** This skill reviews code, dependencies, CI workflows and IaC in a\nrepository. Agent behaviour, tool and MCP permissions, and read-only audits of\nGitHub orgs, DNS or hosts belong to `agent-security-posture`, if it is installed.\n\n## Modes\n\n| Invocation | Scope | Scanner call | `mode` |\n| --- | --- | --- | --- |\n| `\u002Fsecurity-review` | working tree vs HEAD, untracked files included | `--diff HEAD` | `diff` |\n| `\u002Fsecurity-review --base \u003Cref>` | merge-base(ref, HEAD) to the working tree | `--diff $(git merge-base \u003Cref> HEAD)` | `diff` |\n| `\u002Fsecurity-review --pr \u003Cn>` | GitHub PR `n` (check out its head with `gh pr checkout \u003Cn>` first) | `--diff \u003Cpr base sha>` | `pr` |\n| `\u002Fsecurity-review --repo` | whole tree, plus full git history for secrets | `--audit` | `repo` |\n\n## Rules that do not bend\n\n- **Never print a secret.** Report the file, the line and the secret type only:\n  never the value, a fragment of it, or a hash. The scanner script already\n  redacts at capture. Do not undo that by quoting the source line.\n- **Missing is not clean.** A scanner that is not installed, times out or fails\n  goes in `not_inspected`. Exit 2 from either script means unchecked, not safe.\n- **Evidence and judgement are separate.** Scanner output is evidence. The\n  verdict is yours, and you must be able to defend it with the six elements below.\n- **Read-only.** Do not fix, commit or push unless the user asks. Do not\n  install scanners without approval. Report each missing one, with its install\n  command.\n\n## Workflow\n\n### 1. Threat model (STRIDE-lite)\n\nRead `references\u002Fsecurity-checklist.md` and `references\u002Fowasp-top10-2025.md`.\n\nWrite five short lines in your notes before you read any code. Each line names\nsomething specific to this change. Do not pad them with generic advice.\n\n- **Assets**: data, credentials or capabilities this code touches.\n- **Entry points**: HTTP handlers, CLI args, files, queues, webhooks, model\n  output, CI events (`pull_request_target`, `issue_comment`).\n- **Trust boundaries**: where lower-trust input meets higher-trust code.\n- **STRIDE**: which of spoofing, tampering, repudiation, information disclosure,\n  denial of service or elevation of privilege are plausible here, and where.\n- **Out of scope**: what you will not look at, and why.\n\n### 2. Run the scanners\n\nRead `references\u002Fsecrets-and-supply-chain.md` if a scanner is missing or a\nsecret lead appears.\n\n```bash\npython3 \u003Cskill-dir>\u002Fscripts\u002Frun_scanners.py --diff \u003Cbase> --json-out leads.json\n```\n\nUse `--audit` instead of `--diff` for `--repo`. Use `--tools a,b` to run a subset.\n\nThe script runs only the installed tools that fit the repository: semgrep (with\nlanguage packs), gitleaks, osv-scanner, trivy, actionlint and zizmor (only if\nthere are workflows), npm audit, pip-audit, govulncheck and cargo-audit (only\nif there are lockfiles). It prints `{status, tools, leads, not_inspected}`.\n\nExit codes: 0 clean, 1 leads, 2 partial, 64 usage error. Copy its\n`not_inspected` list into your findings document unchanged.\n\nIn diff mode, file-level leads are filtered to changed files. Dependency leads\ncover the whole lockfile. Count each one as a finding only if the change adds\nthe package or makes it reachable.\n\n### 3. Triage every lead\n\nRead `references\u002Fcwe-top25.md` to choose the rule id.\n\nA lead is a finding only when you can name all six of these:\n\n1. the **lower-trust principal** (anonymous user, PR author, tenant, model output)\n2. the **input they control**\n3. the **control that should stop it**, and why it fails here\n4. the **trust boundary** crossed\n5. the **principal or resource affected**\n6. a **concrete result** (data read or changed, code run, money moved, service down)\n\nGive each lead one verdict:\n\n- **confirmed**: all six hold, and you traced the path in the code.\n- **needs_validation**: plausible, but one element depends on something you\n  cannot see, such as config or a deployment detail. Severity is capped at P2.\n- **rejected**: an element is missing. Record why in your notes. A rejected\n  lead never becomes a finding.\n\nCommon rejections: test-only code, unreachable code, input that is already\ntrusted, or a dependency whose vulnerable function is not called.\n\nDefence-in-depth suggestions on their own are not vulnerabilities. Report them\nas `nit`, if at all.\n\nWhen `agent-security-posture` is installed, its `references\u002Faudit-method.md`\nhas the full method and the severity anchors. Where possible, have a second\nagent or a fresh pass check confirmed P0\u002FP1 findings. The finder should not be\nits own checker.\n\n### 4. Manual review\n\nScanners miss logic flaws. Read the changed code, plus enough of its callers to\nfollow data across each trust boundary from step 1. Use these references:\n\n- `references\u002Fowasp-top10-2025.md` and `references\u002Fsecurity-checklist.md`: always.\n- `references\u002Fasvs-5.md`: when the change touches authentication, sessions,\n  access control, crypto, file handling or data protection.\n- `references\u002Fllm-and-api.md`: when the code calls an LLM, handles model\n  output, gives an agent tools, or exposes an HTTP API.\n- `references\u002Fgha-hardening.md`: when `.github\u002Fworkflows\u002F` or actions change.\n- `references\u002Fcheat-sheets.md`: for the fix you will suggest.\n\nLook especially for these, because no scanner here finds them:\n\n- missing or wrong authorisation (IDOR, tenant checks)\n- authentication logic, sessions and token lifetimes\n- business-logic abuse\n- server-side request forgery (SSRF) through URL parameters\n- unsafe deserialisation of trusted-looking data\n- race conditions on money or quotas\n- secrets in logs or errors\n- model output that reaches a shell, SQL, HTML or a tool call\n\n### 5. Live advisories\n\nRead `references\u002Fsecrets-and-supply-chain.md`.\n\nRun this when dependencies changed, or when a lead names a CVE or GHSA.\n\n- **If `agent-security-posture` is installed:**\n  `python3 \u003Cits-dir>\u002Fscripts\u002Fcheck_vulns.py lockfile \u003Clockfile> --enrich`.\n  This adds OSV aliases, CISA KEV membership, EPSS and GHSA details. Exit 2\n  means partial.\n- **Otherwise:** use the osv-scanner and native-audit leads from step 2. You\n  can also query `https:\u002F\u002Fapi.osv.dev\u002Fv1\u002Fquery` with the ecosystem, package and\n  version only.\n\nSend only ecosystem, package and version to outside services. Never send\nlockfiles, paths or code.\n\nRaise the severity of a confirmed, reachable dependency finding if it is on\nKEV or has EPSS above 0.1. Never raise one only because it has a high CVSS score.\n\n### 6. Emit findings\n\nRead `references\u002Fcheat-sheets.md` for the `references[]` URLs, and\n`references\u002Fcwe-top25.md` for rule ids. Write one document:\n\n```json\n{\"schema_version\": \"1\", \"producer\": \"security-review\", \"mode\": \"diff\",\n \"summary\": \"One sentence: what you reviewed and what matters most.\",\n \"inspected\": [\"app.py\", \"semgrep\", \"gitleaks\"],\n \"not_inspected\": [{\"what\": \"trivy\", \"reason\": \"not installed\"}],\n \"findings\": [{\n   \"id\": \"SEC-001\", \"severity\": \"P1\", \"axis\": \"security\", \"rule\": \"CWE-89\",\n   \"title\": \"User id interpolated into SQL in get_user\",\n   \"file\": \"app.py\", \"start_line\": 12, \"end_line\": 12,\n   \"rationale\": \"Anonymous caller controls `id`; the f-string bypasses parameterisation; attacker reads any row in users.\",\n   \"evidence\": \"cursor.execute(f\\\"SELECT * FROM users WHERE id = {user_id}\\\")  # semgrep formatted-sql-query\",\n   \"suggestion\": \"cursor.execute(\\\"SELECT * FROM users WHERE id = ?\\\", (user_id,))\",\n   \"references\": [{\"title\": \"OWASP SQL Injection Prevention Cheat Sheet\",\n                   \"url\": \"https:\u002F\u002Fcheatsheetseries.owasp.org\u002Fcheatsheets\u002FSQL_Injection_Prevention_Cheat_Sheet.html\"}],\n   \"confidence\": \"high\", \"status\": \"new\", \"prior_id\": \"\", \"security_verdict\": \"confirmed\"}]}\n```\n\nField rules:\n\n- `axis` is always `security`.\n- `rule` matches `CWE-n`, `CVE-…`, `GHSA-…`, `OSV-…`, `ASInn`, `LLMnn` or `APInn`.\n  Prefer the most specific CWE. Use the advisory id for a dependency.\n- Severity: critical = `P0`, high = `P1`, medium = `P2`, low = `P3`,\n  informational = `nit`.\n- `needs_validation` findings cannot be P0 or P1.\n- `evidence` holds the code or the tool output, with any secret redacted.\n- `rationale` states the six elements in a sentence or two.\n- `inspected` lists the files and tools you actually covered.\n\nValidate the document, then gate on it and convert it to SARIF:\n\n```bash\npython3 \u003Cskill-dir>\u002Fscripts\u002Ffindings.py validate findings.json\npython3 \u003Cskill-dir>\u002Fscripts\u002Ffindings.py gate findings.json          # exit 1: open P0-P3\npython3 \u003Cskill-dir>\u002Fscripts\u002Ffindings.py sarif findings.json > findings.sarif\n```\n\nFix every validation error before you report.\n\nThen give the user a short rendering:\n\n- findings by severity\n- what was not inspected, and why\n- the install commands for missing scanners\n\nDo not report anything else.\n\n## Scanner install hints\n\n`brew install semgrep gitleaks osv-scanner trivy actionlint zizmor`,\n`pipx install pip-audit`, `go install golang.org\u002Fx\u002Fvuln\u002Fcmd\u002Fgovulncheck@latest`,\n`cargo install cargo-audit`.\n\nSuggest these to the user. Do not run them.\n","# Security Review\n\nYou produce **security findings**: confirmed or plausible vulnerabilities, each\ntied to specific lines and a CWE, CVE or GHSA id. Scanners produce **leads**,\nwhich are unreviewed hits. A lead becomes a finding only once you have triaged\nit. The output is a JSON findings document that `scripts\u002Ffindings.py`\nvalidates and gates on.\n\n**Paths:** `\u003Cskill-dir>` is the directory that contains this SKILL.md. Run\nscripts as `python3 \u003Cskill-dir>\u002Fscripts\u002F\u003Cname>.py`. They use only the Python\nstandard library (3.12+).\n\n**Scope.** This skill reviews code, dependencies, CI workflows and IaC in a\nrepository. Agent behaviour, tool and MCP permissions, and read-only audits of\nGitHub orgs, DNS or hosts belong to `agent-security-posture`, if it is installed.\n\n## Modes\n\n| Invocation | Scope | Scanner call | `mode` |\n| --- | --- | --- | --- |\n| `\u002Fsecurity-review` | working tree vs HEAD, untracked files included | `--diff HEAD` | `diff` |\n| `\u002Fsecurity-review --base \u003Cref>` | merge-base(ref, HEAD) to the working tree | `--diff $(git merge-base \u003Cref> HEAD)` | `diff` |\n| `\u002Fsecurity-review --pr \u003Cn>` | GitHub PR `n` (check out its head with `gh pr checkout \u003Cn>` first) | `--diff \u003Cpr base sha>` | `pr` |\n| `\u002Fsecurity-review --repo` | whole tree, plus full git history for secrets | `--audit` | `repo` |\n\n## Rules that do not bend\n\n- **Never print a secret.** Report the file, the line and the secret type only:\n  never the value, a fragment of it, or a hash. The scanner script already\n  redacts at capture. Do not undo that by quoting the source line.\n- **Missing is not clean.** A scanner that is not installed, times out or fails\n  goes in `not_inspected`. Exit 2 from either script means unchecked, not safe.\n- **Evidence and judgement are separate.** Scanner output is evidence. The\n  verdict is yours, and you must be able to defend it with the six elements below.\n- **Read-only.** Do not fix, commit or push unless the user asks. Do not\n  install scanners without approval. Report each missing one, with its install\n  command.\n\n## Workflow\n\n### 1. Threat model (STRIDE-lite)\n\nRead `references\u002Fsecurity-checklist.md` and `references\u002Fowasp-top10-2025.md`.\n\nWrite five short lines in your notes before you read any code. Each line names\nsomething specific to this change. Do not pad them with generic advice.\n\n- **Assets**: data, credentials or capabilities this code touches.\n- **Entry points**: HTTP handlers, CLI args, files, queues, webhooks, model\n  output, CI events (`pull_request_target`, `issue_comment`).\n- **Trust boundaries**: where lower-trust input meets higher-trust code.\n- **STRIDE**: which of spoofing, tampering, repudiation, information disclosure,\n  denial of service or elevation of privilege are plausible here, and where.\n- **Out of scope**: what you will not look at, and why.\n\n### 2. Run the scanners\n\nRead `references\u002Fsecrets-and-supply-chain.md` if a scanner is missing or a\nsecret lead appears.\n\n```bash\npython3 \u003Cskill-dir>\u002Fscripts\u002Frun_scanners.py --diff \u003Cbase> --json-out leads.json\n```\n\nUse `--audit` instead of `--diff` for `--repo`. Use `--tools a,b` to run a subset.\n\nThe script runs only the installed tools that fit the repository: semgrep (with\nlanguage packs), gitleaks, osv-scanner, trivy, actionlint and zizmor (only if\nthere are workflows), npm audit, pip-audit, govulncheck and cargo-audit (only\nif there are lockfiles). It prints `{status, tools, leads, not_inspected}`.\n\nExit codes: 0 clean, 1 leads, 2 partial, 64 usage error. Copy its\n`not_inspected` list into your findings document unchanged.\n\nIn diff mode, file-level leads are filtered to changed files. Dependency leads\ncover the whole lockfile. Count each one as a finding only if the change adds\nthe package or makes it reachable.\n\n### 3. Triage every lead\n\nRead `references\u002Fcwe-top25.md` to choose the rule id.\n\nA lead is a finding only when you can name all six of these:\n\n1. the **lower-trust principal** (anonymous user, PR author, tenant, model output)\n2. the **input they control**\n3. the **control that should stop it**, and why it fails here\n4. the **trust boundary** crossed\n5. the **principal or resource affected**\n6. a **concrete result** (data read or changed, code run, money moved, service down)\n\nGive each lead one verdict:\n\n- **confirmed**: all six hold, and you traced the path in the code.\n- **needs_validation**: plausible, but one element depends on something you\n  cannot see, such as config or a deployment detail. Severity is capped at P2.\n- **rejected**: an element is missing. Record why in your notes. A rejected\n  lead never becomes a finding.\n\nCommon rejections: test-only code, unreachable code, input that is already\ntrusted, or a dependency whose vulnerable function is not called.\n\nDefence-in-depth suggestions on their own are not vulnerabilities. Report them\nas `nit`, if at all.\n\nWhen `agent-security-posture` is installed, its `references\u002Faudit-method.md`\nhas the full method and the severity anchors. Where possible, have a second\nagent or a fresh pass check confirmed P0\u002FP1 findings. The finder should not be\nits own checker.\n\n### 4. Manual review\n\nScanners miss logic flaws. Read the changed code, plus enough of its callers to\nfollow data across each trust boundary from step 1. Use these references:\n\n- `references\u002Fowasp-top10-2025.md` and `references\u002Fsecurity-checklist.md`: always.\n- `references\u002Fasvs-5.md`: when the change touches authentication, sessions,\n  access control, crypto, file handling or data protection.\n- `references\u002Fllm-and-api.md`: when the code calls an LLM, handles model\n  output, gives an agent tools, or exposes an HTTP API.\n- `references\u002Fgha-hardening.md`: when `.github\u002Fworkflows\u002F` or actions change.\n- `references\u002Fcheat-sheets.md`: for the fix you will suggest.\n\nLook especially for these, because no scanner here finds them:\n\n- missing or wrong authorisation (IDOR, tenant checks)\n- authentication logic, sessions and token lifetimes\n- business-logic abuse\n- server-side request forgery (SSRF) through URL parameters\n- unsafe deserialisation of trusted-looking data\n- race conditions on money or quotas\n- secrets in logs or errors\n- model output that reaches a shell, SQL, HTML or a tool call\n\n### 5. Live advisories\n\nRead `references\u002Fsecrets-and-supply-chain.md`.\n\nRun this when dependencies changed, or when a lead names a CVE or GHSA.\n\n- **If `agent-security-posture` is installed:**\n  `python3 \u003Cits-dir>\u002Fscripts\u002Fcheck_vulns.py lockfile \u003Clockfile> --enrich`.\n  This adds OSV aliases, CISA KEV membership, EPSS and GHSA details. Exit 2\n  means partial.\n- **Otherwise:** use the osv-scanner and native-audit leads from step 2. You\n  can also query `https:\u002F\u002Fapi.osv.dev\u002Fv1\u002Fquery` with the ecosystem, package and\n  version only.\n\nSend only ecosystem, package and version to outside services. Never send\nlockfiles, paths or code.\n\nRaise the severity of a confirmed, reachable dependency finding if it is on\nKEV or has EPSS above 0.1. Never raise one only because it has a high CVSS score.\n\n### 6. Emit findings\n\nRead `references\u002Fcheat-sheets.md` for the `references[]` URLs, and\n`references\u002Fcwe-top25.md` for rule ids. Write one document:\n\n```json\n{\"schema_version\": \"1\", \"producer\": \"security-review\", \"mode\": \"diff\",\n \"summary\": \"One sentence: what you reviewed and what matters most.\",\n \"inspected\": [\"app.py\", \"semgrep\", \"gitleaks\"],\n \"not_inspected\": [{\"what\": \"trivy\", \"reason\": \"not installed\"}],\n \"findings\": [{\n   \"id\": \"SEC-001\", \"severity\": \"P1\", \"axis\": \"security\", \"rule\": \"CWE-89\",\n   \"title\": \"User id interpolated into SQL in get_user\",\n   \"file\": \"app.py\", \"start_line\": 12, \"end_line\": 12,\n   \"rationale\": \"Anonymous caller controls `id`; the f-string bypasses parameterisation; attacker reads any row in users.\",\n   \"evidence\": \"cursor.execute(f\\\"SELECT * FROM users WHERE id = {user_id}\\\")  # semgrep formatted-sql-query\",\n   \"suggestion\": \"cursor.execute(\\\"SELECT * FROM users WHERE id = ?\\\", (user_id,))\",\n   \"references\": [{\"title\": \"OWASP SQL Injection Prevention Cheat Sheet\",\n                   \"url\": \"https:\u002F\u002Fcheatsheetseries.owasp.org\u002Fcheatsheets\u002FSQL_Injection_Prevention_Cheat_Sheet.html\"}],\n   \"confidence\": \"high\", \"status\": \"new\", \"prior_id\": \"\", \"security_verdict\": \"confirmed\"}]}\n```\n\nField rules:\n\n- `axis` is always `security`.\n- `rule` matches `CWE-n`, `CVE-…`, `GHSA-…`, `OSV-…`, `ASInn`, `LLMnn` or `APInn`.\n  Prefer the most specific CWE. Use the advisory id for a dependency.\n- Severity: critical = `P0`, high = `P1`, medium = `P2`, low = `P3`,\n  informational = `nit`.\n- `needs_validation` findings cannot be P0 or P1.\n- `evidence` holds the code or the tool output, with any secret redacted.\n- `rationale` states the six elements in a sentence or two.\n- `inspected` lists the files and tools you actually covered.\n\nValidate the document, then gate on it and convert it to SARIF:\n\n```bash\npython3 \u003Cskill-dir>\u002Fscripts\u002Ffindings.py validate findings.json\npython3 \u003Cskill-dir>\u002Fscripts\u002Ffindings.py gate findings.json          # exit 1: open P0-P3\npython3 \u003Cskill-dir>\u002Fscripts\u002Ffindings.py sarif findings.json > findings.sarif\n```\n\nFix every validation error before you report.\n\nThen give the user a short rendering:\n\n- findings by severity\n- what was not inspected, and why\n- the install commands for missing scanners\n\nDo not report anything else.\n\n## Scanner install hints\n\n`brew install semgrep gitleaks osv-scanner trivy actionlint zizmor`,\n`pipx install pip-audit`, `go install golang.org\u002Fx\u002Fvuln\u002Fcmd\u002Fgovulncheck@latest`,\n`cargo install cargo-audit`.\n\nSuggest these to the user. Do not run them.\n",{"name":5,"license":15,"description":7},"main","f752f2b5b32c4d11b09266675bb60d2c1e6e7680","skills\u002Fsecurity-review",1,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge","https:\u002F\u002Fgithub.com\u002Fksprashu","http:\u002F\u002Fwww.prashu.com","Leading Google Cloud Developer Relations (Advocacy) for JAPAC, focussing on GenAI, AI Agents and AI Assisted Engineering","2026-11-14 21:40:45",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add ksprashu\u002Fagent-skill-forge --skill security-review",{"tool":39,"command":40},"bunx","bunx skills add ksprashu\u002Fagent-skill-forge --skill security-review",{"tool":42,"command":43},"pnpm","pnpm dlx skills add ksprashu\u002Fagent-skill-forge --skill security-review","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002FSKILL.md",true,[47,53,57,61,65,69,73,77,81,85,89,95,99,103,107,111,117,121,125,129,133,137,141,145,149,153,157],{"path":48,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":51,"sourceHtmlUrl":52},"DESIGN.md","REFERENCE","参考文档",4024,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002FDESIGN.md",{"path":54,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":55,"sourceHtmlUrl":56},"references\u002Fasvs-5.md",7269,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Freferences\u002Fasvs-5.md",{"path":58,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":59,"sourceHtmlUrl":60},"references\u002Fcheat-sheets.md",11597,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Freferences\u002Fcheat-sheets.md",{"path":62,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":63,"sourceHtmlUrl":64},"references\u002Fcwe-top25.md",5275,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Freferences\u002Fcwe-top25.md",{"path":66,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":67,"sourceHtmlUrl":68},"references\u002Ffindings.schema.json",4916,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Freferences\u002Ffindings.schema.json",{"path":70,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":71,"sourceHtmlUrl":72},"references\u002Fgha-hardening.md",9318,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Freferences\u002Fgha-hardening.md",{"path":74,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":75,"sourceHtmlUrl":76},"references\u002Fllm-and-api.md",8683,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Freferences\u002Fllm-and-api.md",{"path":78,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":79,"sourceHtmlUrl":80},"references\u002Fowasp-top10-2025.md",12822,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Freferences\u002Fowasp-top10-2025.md",{"path":82,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":83,"sourceHtmlUrl":84},"references\u002Fsecrets-and-supply-chain.md",10025,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Freferences\u002Fsecrets-and-supply-chain.md",{"path":86,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":87,"sourceHtmlUrl":88},"references\u002Fsecurity-checklist.md",22392,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Freferences\u002Fsecurity-checklist.md",{"path":90,"resourceType":91,"resourceTypeDesc":92,"sizeBytes":93,"sourceHtmlUrl":94},"scripts\u002Ffindings.py","SCRIPT","脚本",16243,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Fscripts\u002Ffindings.py",{"path":96,"resourceType":91,"resourceTypeDesc":92,"sizeBytes":97,"sourceHtmlUrl":98},"scripts\u002Frun_scanners.py",38745,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Fscripts\u002Frun_scanners.py",{"path":100,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":101,"sourceHtmlUrl":102},"tests\u002Ffixtures\u002Fcanned\u002Factionlint.json",569,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Ftests\u002Ffixtures\u002Fcanned\u002Factionlint.json",{"path":104,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":105,"sourceHtmlUrl":106},"tests\u002Ffixtures\u002Fcanned\u002Fcargo-audit.json",1181,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Ftests\u002Ffixtures\u002Fcanned\u002Fcargo-audit.json",{"path":108,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":109,"sourceHtmlUrl":110},"tests\u002Ffixtures\u002Fcanned\u002Fgitleaks.json",1146,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Ftests\u002Ffixtures\u002Fcanned\u002Fgitleaks.json",{"path":112,"resourceType":113,"resourceTypeDesc":114,"sizeBytes":115,"sourceHtmlUrl":116},"tests\u002Ffixtures\u002Fcanned\u002Fgovulncheck.jsonl","OTHER","其他",2006,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Ftests\u002Ffixtures\u002Fcanned\u002Fgovulncheck.jsonl",{"path":118,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":119,"sourceHtmlUrl":120},"tests\u002Ffixtures\u002Fcanned\u002Fnpm-audit.json",4814,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Ftests\u002Ffixtures\u002Fcanned\u002Fnpm-audit.json",{"path":122,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":123,"sourceHtmlUrl":124},"tests\u002Ffixtures\u002Fcanned\u002Fosv-scanner.json",6099,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Ftests\u002Ffixtures\u002Fcanned\u002Fosv-scanner.json",{"path":126,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":127,"sourceHtmlUrl":128},"tests\u002Ffixtures\u002Fcanned\u002Fpip-audit.json",6916,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Ftests\u002Ffixtures\u002Fcanned\u002Fpip-audit.json",{"path":130,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":131,"sourceHtmlUrl":132},"tests\u002Ffixtures\u002Fcanned\u002Fsemgrep.json",10796,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Ftests\u002Ffixtures\u002Fcanned\u002Fsemgrep.json",{"path":134,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":135,"sourceHtmlUrl":136},"tests\u002Ffixtures\u002Fcanned\u002Ftrivy.json",4271,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Ftests\u002Ffixtures\u002Fcanned\u002Ftrivy.json",{"path":138,"resourceType":49,"resourceTypeDesc":50,"sizeBytes":139,"sourceHtmlUrl":140},"tests\u002Ffixtures\u002Fcanned\u002Fzizmor.json",10718,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Ftests\u002Ffixtures\u002Fcanned\u002Fzizmor.json",{"path":142,"resourceType":91,"resourceTypeDesc":92,"sizeBytes":143,"sourceHtmlUrl":144},"tests\u002Ffixtures\u002Frepos\u002Fclean\u002Fapp.py",385,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Ftests\u002Ffixtures\u002Frepos\u002Fclean\u002Fapp.py",{"path":146,"resourceType":91,"resourceTypeDesc":92,"sizeBytes":147,"sourceHtmlUrl":148},"tests\u002Ffixtures\u002Frepos\u002Fclean\u002Fconfig.py",14,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Ftests\u002Ffixtures\u002Frepos\u002Fclean\u002Fconfig.py",{"path":150,"resourceType":91,"resourceTypeDesc":92,"sizeBytes":151,"sourceHtmlUrl":152},"tests\u002Ffixtures\u002Frepos\u002Fvulnerable\u002Fapp.py",553,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Ftests\u002Ffixtures\u002Frepos\u002Fvulnerable\u002Fapp.py",{"path":154,"resourceType":91,"resourceTypeDesc":92,"sizeBytes":155,"sourceHtmlUrl":156},"tests\u002Ftest_security_review_normalisers.py",10745,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Ftests\u002Ftest_security_review_normalisers.py",{"path":158,"resourceType":91,"resourceTypeDesc":92,"sizeBytes":159,"sourceHtmlUrl":160},"tests\u002Ftest_security_review_run.py",17704,"https:\u002F\u002Fgithub.com\u002Fksprashu\u002Fagent-skill-forge\u002Fblob\u002Fmain\u002Fskills\u002Fsecurity-review\u002Ftests\u002Ftest_security_review_run.py",[]]