docs(credits): recognize contributors from PRs #219, #220, #239-#240, #257-#260

CHANGELOG.md

@@ -28,6 +28,12 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   is still reported, as its own signal rather than as a change, through
   the `authority` block per-capability `provenance_class`. Gate outcomes,
   severity capping, and exit codes are unchanged.
+  Thanks to [@tyza66](https://github.com/tyza66) (PR #257).
+- `docs/github-actions-example.md` documents the workflow trust boundary:
+  `contents: read` for the minimal scan, the extra `pull-requests: write`
+  needed only for `pr-comment-post`, and keeping broad tokens out of
+  `extra-args` and the environment.
+  Thanks to [@HectorGuerilla](https://github.com/HectorGuerilla) (PR #259).
 
 ### Fixed
 
@@ -39,6 +45,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   stderr naming the gates that cannot fail this run. Exit codes are
   unchanged. `--unknown-authority pass` and `review` cannot fail a run,
   so they are not named.
+  Thanks to [@tyza66](https://github.com/tyza66) (PR #258).
 - That warning now decides the unknown-authority gate from the
   *effective* policy — a policy file or profile setting
   `authority.unknown: deny` is the same gate as
@@ -48,6 +55,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   alongside the flag gates, an explicit `allow`/`require_review` setting
   wins over the flag and is not named, and the warning is labelled by
   whichever source put `deny` in force. Exit codes are unchanged.
+  Thanks to [@tyza66](https://github.com/tyza66) (PR #260).
 
 ## [2.7.0] - 2026-10-22
 

CONTRIBUTING.md

@@ -151,7 +151,7 @@ We are grateful to the following community members for their contributions:
 | Contributor | Contributions |
 |-------------|---------------|
 | [@i-safonoff](https://github.com/i-safonoff) | `.cursorrules` framework adapter, `rule_coverage_summary()`, RULES_REFERENCE.md, dataflow rule ID casing fix |
-| [@ARAVIND281](https://github.com/ARAVIND281) | Claude Code permission evaluation order fix, interprocedural data-flow tracking |
+| [@ARAVIND281](https://github.com/ARAVIND281) | Claude Code permission evaluation order fix, interprocedural data-flow tracking, exception scope-escape boundary tests (PR #220) |
 | [@Solarthis](https://github.com/Solarthis) | MCP tool description injection detection |
 | [@Aming9303](https://github.com/Aming9303) | `safeai registry components` CLI, `safeai init` command, GitHub Actions workflow example |
 | [@adnqcr7-code](https://github.com/adnqcr7-code) | Framework detection tests (LangGraph, CrewAI, LlamaIndex, n8n, Claude Code), CI integration docs, SARIF review guidance |
@@ -170,6 +170,10 @@ We are grateful to the following community members for their contributions:
 | [@chenzeyan54-commits](https://github.com/chenzeyan54-commits) | Terminal output sanitization against ANSI/OSC injection (PR #203) |
 | [@HarshRajSinghania](https://github.com/HarshRajSinghania) | PR comment markdown injection sanitization (PR #204) |
 | [@thadidaniel-ctrl](https://github.com/thadidaniel-ctrl) | Release workflow supply-chain posture tests (PR #206) |
+| [@Golden007-prog](https://github.com/Golden007-prog) | Authority Evidence Benchmark HTML presentation, history limit (PRs #239, #240) |
+| [@therealestbenji](https://github.com/therealestbenji) | Microsoft Teams integration capability detector `CAP_ms_teams` (PR #219) |
+| [@tyza66](https://github.com/tyza66) | Authority-gate fail-closed audit: per-side access-mode provenance, diff-gate blind warning covering flag and policy-file gates (PRs #257, #258, #260) |
+| [@HectorGuerilla](https://github.com/HectorGuerilla) | GitHub Actions trust-boundary and permissions guidance (PR #259) |
 
 Your contributions help make AI safer for everyone. Thank you!
 

README.md

@@ -557,7 +557,7 @@ contributors who have helped make AI safer:
 | Contributor | Key Contributions |
 |-------------|-------------------|
 | [@i-safonoff](https://github.com/i-safonoff) | `.cursorrules` framework adapter, `rule_coverage_summary()`, RULES_REFERENCE.md, dataflow casing fix |
-| [@ARAVIND281](https://github.com/ARAVIND281) | Claude Code permission evaluation order, interprocedural data-flow tracking |
+| [@ARAVIND281](https://github.com/ARAVIND281) | Claude Code permission evaluation order, interprocedural data-flow tracking, exception scope-escape boundary tests (PR #220) |
 | [@Solarthis](https://github.com/Solarthis) | MCP tool description injection detection |
 | [@Aming9303](https://github.com/Aming9303) | `safeai registry components` CLI, `safeai init` command, GitHub Actions example |
 | [@adnqcr7-code](https://github.com/adnqcr7-code) | Framework detection tests (LangGraph, CrewAI, LlamaIndex, n8n, Claude Code), CI/SARIF docs |
@@ -577,6 +577,9 @@ contributors who have helped make AI safer:
 | [@HarshRajSinghania](https://github.com/HarshRajSinghania) | PR comment markdown injection sanitization (PR #204) |
 | [@thadidaniel-ctrl](https://github.com/thadidaniel-ctrl) | Release workflow supply-chain posture tests (PR #206) |
 | [@Golden007-prog](https://github.com/Golden007-prog) | Authority Evidence Benchmark HTML presentation, history limit (PRs #239, #240) |
+| [@therealestbenji](https://github.com/therealestbenji) | Microsoft Teams integration capability detector `CAP_ms_teams` (PR #219) |
+| [@tyza66](https://github.com/tyza66) | Authority-gate fail-closed audit: per-side access-mode provenance, diff-gate blind warning covering flag and policy-file gates (PRs #257, #258, #260) |
+| [@HectorGuerilla](https://github.com/HectorGuerilla) | GitHub Actions trust-boundary and permissions guidance (PR #259) |
 
 See [CONTRIBUTING.md](./CONTRIBUTING.md) for how to get involved.
 

docs/contributing/GOOD_FIRST_ISSUES.md

@@ -35,6 +35,7 @@ contributors. See the [Completed Issues](#-completed-issues) section at the bott
 - **Difficulty:** Easy | **Effort:** 2–3 hours
 - **Suggested files:** `safeai/analyzers/capability/analyzer.py`, `safeai/analysis/capabilities.py`
 - **Description:** Add a regex pattern to detect Microsoft Teams SDK usage (`import teams`, `from teams import`, `TeamsClient`, etc.). Follow the same pattern as the existing Docker, Slack, and Jira detectors.
+- **Status:** ✅ Complete — `CAP_ms_teams` rule, detector and tests (@therealestbenji, PR #219)
 
 ### 4. Detect SharePoint access
 - **Difficulty:** Easy | **Effort:** 2 hours
@@ -195,12 +196,14 @@ Newest batch — issue-driven security review and test-hardening tasks for the v
 - **Labels:** good first issue, security
 - **Suggested files:** `safeai/kya/exceptions.py`, `tests/test_exceptions.py`
 - **Description:** Verify traversal-style `target_id` values (`../`, absolute paths, Windows `..\\`), legacy-key ambiguity, and cross-project exceptions can never resolve to `active`. Fail-closed: unknown scope = not applied.
+- **Status:** ✅ Complete — scope-escape boundaries pinned, unverified scope gated (@ARAVIND281, PR #220)
 
 ### 55. Security review: authority gate fail-closed guarantees (#197)
 - **Difficulty:** Easy-Medium | **Effort:** 2–3 hours
 - **Labels:** good first issue, security
 - **Suggested files:** `safeai/analysis/capability_diff.py`, `safeai/cmd/postprocess.py`, `tests/test_authority_change.py`, `tests/test_unknown_authority.py`
 - **Description:** Lock non-zero exit for: confirmed authority change with `--fail-on-authority-change`, UNKNOWN authority under `block`, and mixed inferred/confirmed additions (`inferred_only` must not mask confirmed tools).
+- **Status:** In progress — per-side access-mode provenance and the diff-gate blind warning landed (@tyza66, PRs #257, #258, #260); `--unknown-authority block` without `--baseline` and removed-tool UNKNOWN handling remain open on #197
 
 ### 56. Security test: HTML report injection fixtures (#198)
 - **Difficulty:** Easy | **Effort:** 2 hours
@@ -262,6 +265,7 @@ These issues have been implemented by community contributors and are now part of
 | 18 | Detect Jira integration | @yugaaank (PR #2) |
 | 19 | Detect browser automation capability | @yugaaank (PR #2) |
 | — | Detect hidden instructions in MCP tool descriptions | @Solarthis (PR #107) |
+| #182 | Detect Microsoft Teams integration | @therealestbenji (PR #219) |
 
 ### Reports, Output & Scoring
 
@@ -303,6 +307,7 @@ These issues have been implemented by community contributors and are now part of
 
 | # | Issue | Contributor |
 |---|-------|-------------|
+| #196 | Exception scope-escape paths | @ARAVIND281 (PR #220) |
 | #199 | Terminal escape-sequence sanitization | @chenzeyan54-commits (PR #203) |
 | #200 | PR comment markdown injection sanitization | @HarshRajSinghania (PR #204) |
 | #201 | Release workflow supply-chain posture tests | @thadidaniel-ctrl (PR #206) |