1. SECRET_PATTERN had [A-Za-z0-9+/_\-!@#] -- backslash-escaped hyphen
is invalid POSIX ERE; grep parsed it as a range with invalid bounds.
Fix: move hyphen to end of class: [A-Za-z0-9+/_!@#-].
2. printf -- '---\n' fails with 'invalid option' in bash because the
builtin does not accept -- as end-of-options. Removed -- from all
four printf calls.
Shell heredocs with unindented bodies (line 1 content) terminate YAML
run: | block scalars. The YAML parser sees the unindented heredoc body
as leaving the block, making the workflow file unparseable -- Gitea
silently stops creating runs for a workflow with invalid YAML.
Replace the single-quoted heredoc prompt with a group of printf + cat
calls. Every line stays properly indented within the YAML block scalar.
Use jq --rawfile instead of --arg to load the prompt from a temp file,
which also eliminates shell escaping hazards for large strings.
Gitea 1.22 cancel-in-progress does not behave like GitHub Actions: when
a new synchronize event arrives while a review is running, instead of
cancelling the running job and starting a new one, it drops the new run
silently. Remove the concurrency block entirely so every commit to a PR
gets its own review run.
The PROMPT string contained backtick-quoted text for the Evidence field
example. Inside a double-quoted bash string, backticks trigger command
substitution, causing 'exact: command not found' at runtime.
Fix: build the prompt using a single-quoted heredoc (no shell expansion
inside) then splice dynamic values via sed and python3 replace() instead
of shell variable interpolation.
Two changes to reduce hallucinations in pr-review:
1. Codebase index (new step "Build codebase index"):
Generates a compact manifest of everything that EXISTS in the project:
- All registered Tauri commands (from lib.rs generate_handler![])
- All TypeScript exports (from tauriCommands.ts)
- All public Rust fn signatures in commands/
- All DB migration names
This index is prepended to the prompt so the model cannot invent
functions like authenticate_sudo or continue_chat_history that are
absent from both the index and the file contents.
2. Full-repo verification (updated "Verify findings" step):
Previously only grepped changed files, which falsely tagged findings
about unchanged-but-real code as UNVERIFIED. Now runs git ls-files
to load all tracked source files, so verification only fails for
code that genuinely does not exist anywhere in the codebase.
If qwen3-coder continues to hallucinate after these changes, swap the
model name on line 184 to bedrock-personal or claude-haiku.
qwen3-coder-next fabricates plausible-looking code in its Evidence
blocks instead of quoting from the actual files provided. This adds a
Python verification step that greps each fenced code block against the
real changed files and tags any finding whose evidence cannot be found
as UNVERIFIED.
This is a safeguard, not a fix — the model is fundamentally unreliable
for grounded code review. The longer-term fix is to replace qwen3-coder
with a model that stays grounded to context (Claude Haiku, devstral,
or deepseek-coder-v2 via the LiteLLM proxy / vLLM at 172.0.1.42).
The previous regex matched any line containing "password", "token", etc.
near certain punctuation characters. This silently removed function
signatures, variable declarations, and test assertions from the context
sent to the LLM — causing it to hallucinate 3 BLOCKERs per review:
- "function signature missing" (the `password: &str` param was scrubbed)
- "filter body empty" (the filter condition containing "password" was scrubbed)
- "password passed unencrypted" (the decrypt_token call line was scrubbed)
Fix: match actual credential VALUES only:
- Well-known token formats (AKIA..., ghp_..., xox...)
- keyword = "long_quoted_literal" (25+ chars, clearly a value not a name)
- Standalone base64 blob lines (60+ chars, PEM-style)
Never scrub a line just because it contains a credential-related word.
Rust 1.88 enforces clippy::uninlined_format_args as a style lint under
-D warnings. Change `writeln!(stdin, "{}", password)` to the inline
form `writeln!(stdin, "{password}")`.
Three changes:
- Exclude Cargo.lock/lockfiles from the diff — removes ~163 lines of
hash noise that waste the review budget with no value
- Raise line cap from 500 to 3000 and add a truncation notice when
the diff is cut, so the model knows the diff is incomplete
- Harden prompt: require quoted evidence for every finding; add explicit
self-verification step for missing-identifier claims (search full diff
before raising); tighten no-hallucinate instruction
Only a single hardcoded entry (word/document.xml) is ever accessed from
the ZIP archive; no arbitrary path extraction occurs, so path traversal
attacks cannot apply. Add a comment to make this invariant explicit for
future maintainers.
- Add extension allowlist (SAFE_TEXT_EXTENSIONS + SAFE_BINARY_EXTENSIONS)
rejecting unsupported file types at both upload_log_file and
upload_log_file_by_content entry points
- Add extract_text_content() with PDF text extraction via lopdf and
DOCX extraction via zip+quick-xml
- Binary files (PDF/DOCX) get extracted text written to .extracted.txt
for downstream PII detection
- Expand frontend file input accept list and add collapsible
supported-formats disclosure element
- Add 11 unit tests covering allowlist logic and extraction paths
AI history continuity: Changed the history-load query in chat_message to
JOIN ai_conversations and select by issue_id instead of single conversation_id.
This preserves full context when provider/model changes mid-triage.
Deep search: Added DISTINCT to list_issues SELECT and extended the search
filter with EXISTS subqueries covering ai_messages, resolution_steps,
log_files, and timeline_events. Ensures comprehensive search without
duplicate results.
Includes 11 new unit tests covering both features.
Cargo.toml was updated to 0.3.0 for the MCP feature but tauri.conf.json
was missed. Bundle artifact filenames are derived from tauri.conf.json,
so all builds were producing files named 0.2.68.
tea is not available in alpine:latest containers. Replace both tea calls
with curl API requests against the Gitea releases endpoint.
Also: auto-tag now reads src-tauri/Cargo.toml version first. If it is
ahead of the latest git tag (major/minor bump), that version is used
directly instead of incrementing the patch.
- call_tool: 30s hard timeout via tokio::time::timeout
- discover_server: 60s hard timeout wrapping full connect+discover sequence
- delete_mcp_server: write_audit_event before cascade delete, capturing
server name, tool count, and resource count
- initiate_mcp_oauth: append cryptographically random state nonce for CSRF
- pr-review.yml: rewrite prompt to require line-quoted evidence for every
finding and eliminate hallucinated false positives
- Implement AgentRegistry system with devops-incident-responder agent
- Add domain detection based on conversation keywords
- Inject devops-incident-responder as primary system prompt
- Auto-switch domain prompts silently when context shifts
- Fix version update script to handle JSON format correctly
- Always display version in bottom-left corner
- Add release notes fallback to git commits if CHANGELOG empty
This implements the full devops-incident-responder agent as the primary
system prompt, with domain-specific SME prompts layered on top based on
conversation content analysis. The version display bug is fixed by removing
the collapsed condition, and release notes now have a fallback mechanism.
Replace direct Ollama API calls with liteLLM proxy at
172.0.0.29:11434 using qwen2.5-72b (72B VLLM model). Increase
timeouts to 300s for larger model inference. Reuses existing
OLLAMA_API_KEY secret for liteLLM auth.
Also add push-to-master trigger on test.yml so merges to master
run the full CI suite (previously only pull_request events triggered).
