fix: replace hardcoded user paths with tilde notation in Claude settings#6
Closed
jankowtf wants to merge 1 commit into
Closed
fix: replace hardcoded user paths with tilde notation in Claude settings#6jankowtf wants to merge 1 commit into
jankowtf wants to merge 1 commit into
Conversation
🤖 Generated with [Claude Code](https://claude.ai/code) Co-Authored-By: Claude <noreply@anthropic.com>
Author
|
Extended it and new PR |
This was referenced Feb 10, 2026
rikitikitavi2012-debug
added a commit
to rikitikitavi2012-debug/PAI
that referenced
this pull request
Mar 10, 2026
Added comprehensive tests for `SecurityValidator.hook.ts`, `ModeClassifier.hook.ts`, `PreCompact.hook.ts`, and `LoadContext.hook.ts`. Addressed testing edge cases such as malformed JSON parsing, empty input gracefully passing (fail-open), and explicitly verifying <500ms execution times as part of the assertions. Co-authored-by: google-labs-jules[bot] <161369871+google-labs-jules[bot]@users.noreply.github.com> Co-authored-by: rikitikitavi2012-debug <240362902+rikitikitavi2012-debug@users.noreply.github.com>
rikitikitavi2012-debug
added a commit
to rikitikitavi2012-debug/PAI
that referenced
this pull request
Mar 30, 2026
Cross-model finding danielmiessler#6: ModeClassifier/Complexity Gate/ISC Gate conflated. Added explicit mechanism chain showing 3 independent levels. Rebuilt CLAUDE.md from updated template.
rikitikitavi2012-debug
added a commit
to rikitikitavi2012-debug/PAI
that referenced
this pull request
Apr 2, 2026
Completed: - #4 Hook filtering - danielmiessler#15 showThinkingSummaries - danielmiessler#12 MCP_CONNECTION_NONBLOCKING - danielmiessler#6 EVIDENCE_FIRST gate - danielmiessler#16 PhaseGate hook - danielmiessler#9 Circuit Breaker Skipped: danielmiessler#10 (use Anthropic API + Z.AI directly)
Alessandro-Improta
added a commit
to Alessandro-Improta/Personal_AI_Infrastructure
that referenced
this pull request
May 1, 2026
…ator, and tools danielmiessler#2 — reloadPulse() (actions.ts): removed early-return that checked for ~/Library/LaunchAgents/com.pai.pulse.plist before attempting restart. On Linux that path never exists, so voice config was never applied after install. Now branches on process.platform: checks the systemd unit file (~/.config/systemd/user/pai-pulse.service) on Linux, plist on macOS. danielmiessler#3 — Post-install validator (validate.ts): - "Pulse auto-start" check now resolves the correct supervisor path per platform (systemd service on Linux, launchd plist on macOS) and names the check accordingly ("Pulse systemd service" / "Pulse launchd agent"). - Shell alias check now scans .zshrc, .bashrc, and .profile in order and reports which file the alias was found in, so Ubuntu bash users no longer see a false CRITICAL failure after a clean install. danielmiessler#4 — start-pulse.sh: gated the `security unlock-keychain` call behind `[ "$(uname -s)" = "Darwin" ]`. The `security` binary does not exist on Linux; the 2>/dev/null suppressed the error but the intent (keychain unlock) was meaningless on Linux anyway. danielmiessler#5 — pai.ts: wrapped the osascript wallpaper block in `process.platform === "darwin"`. Was unguarded; on Linux spawnSync would log a noisy warning on every wallpaper change regardless of outcome. danielmiessler#6 — PreviewMarkdown.ts: replaced unconditional `open` (macOS-only) with a platform-aware `opener` (open on darwin, xdg-open on Linux). Added .catch(() => {}) so headless Linux environments (no DISPLAY) fail silently rather than throwing.
neilinger
added a commit
to neilinger/Personal_AI_Infrastructure
that referenced
this pull request
May 3, 2026
Addresses code-review findings post-installer-refactor: danielmiessler#3 Mirror-lib drift guard: hooks/lib/paths.test.ts adds a unit test that diffs hooks/lib/paths.ts against PAI/lib/paths.ts (post-docblock). The two are deliberate mirrors per C1; this test fails the suite if they drift. No CI plumbing needed — runs in the regular bun test pass. danielmiessler#4 engine/paths.ts:53 clean() consistency: replaced direct process.env.PAI_DIR.trim() check with a clean() helper invocation matching the rest of the function. Same semantics, explicit intent, no future drift if clean() ever changes. danielmiessler#5 paths.sh exit-vs-return: documented the "return 1 2>/dev/null || exit 1" pattern with a four-line comment. danielmiessler#7 Test gap closed: new rows in PAI-Install/engine/paths.test.ts cover cliClaude + PAI_DIR=whitespace and cliClaude + PAI_DIR=empty. (Was implicit before; now explicit.) 41 tests total (was 38). danielmiessler#6 Untracked dirs: added .claude/worktrees/ and .omc/ to superproject gitignore; added .omc/ to release-tree gitignore. Architecture references: switched from "L18-34" to '"## Directory Structure"' (section name, stable across edits) in hooks/lib/paths.ts, PAI/lib/paths.ts, PAI/lib/paths.sh, PAI/lib/paths.bats, hooks/lib/paths.test.ts, PAI/PAI-Install/engine/paths.ts, PAI/PAI-Install/engine/paths.test.ts, all 3 task YAMLs and the suite YAML. Old eval Results/ JSONs left as historical run artifacts. Verification: 41/41 TS GREEN (17 installer + 24 path-lib including new drift guard); 12/12 bats GREEN; shell helper smoke clean. Not addressed in this commit (per review): - #1 rebase onto origin/main (your call to push) - #2 settings.json/CLAUDE.md local overrides (left uncommitted)
neilinger
added a commit
to neilinger/Personal_AI_Infrastructure
that referenced
this pull request
May 3, 2026
Addresses code-review findings post-installer-refactor: danielmiessler#3 Mirror-lib drift guard: hooks/lib/paths.test.ts adds a unit test that diffs hooks/lib/paths.ts against PAI/lib/paths.ts (post-docblock). The two are deliberate mirrors per C1; this test fails the suite if they drift. No CI plumbing needed — runs in the regular bun test pass. danielmiessler#4 engine/paths.ts:53 clean() consistency: replaced direct process.env.PAI_DIR.trim() check with a clean() helper invocation matching the rest of the function. Same semantics, explicit intent, no future drift if clean() ever changes. danielmiessler#5 paths.sh exit-vs-return: documented the "return 1 2>/dev/null || exit 1" pattern with a four-line comment. danielmiessler#7 Test gap closed: new rows in PAI-Install/engine/paths.test.ts cover cliClaude + PAI_DIR=whitespace and cliClaude + PAI_DIR=empty. (Was implicit before; now explicit.) 41 tests total (was 38). danielmiessler#6 Untracked dirs: added .claude/worktrees/ and .omc/ to superproject gitignore; added .omc/ to release-tree gitignore. Architecture references: switched from "L18-34" to '"## Directory Structure"' (section name, stable across edits) in hooks/lib/paths.ts, PAI/lib/paths.ts, PAI/lib/paths.sh, PAI/lib/paths.bats, hooks/lib/paths.test.ts, PAI/PAI-Install/engine/paths.ts, PAI/PAI-Install/engine/paths.test.ts, all 3 task YAMLs and the suite YAML. Old eval Results/ JSONs left as historical run artifacts. Verification: 41/41 TS GREEN (17 installer + 24 path-lib including new drift guard); 12/12 bats GREEN; shell helper smoke clean. Not addressed in this commit (per review): - #1 rebase onto origin/main (your call to push) - #2 settings.json/CLAUDE.md local overrides (left uncommitted)
neilinger
added a commit
to neilinger/Personal_AI_Infrastructure
that referenced
this pull request
May 3, 2026
Addresses code-review findings post-installer-refactor: danielmiessler#3 Mirror-lib drift guard: hooks/lib/paths.test.ts adds a unit test that diffs hooks/lib/paths.ts against PAI/lib/paths.ts (post-docblock). The two are deliberate mirrors per C1; this test fails the suite if they drift. No CI plumbing needed — runs in the regular bun test pass. danielmiessler#4 engine/paths.ts:53 clean() consistency: replaced direct process.env.PAI_DIR.trim() check with a clean() helper invocation matching the rest of the function. Same semantics, explicit intent, no future drift if clean() ever changes. danielmiessler#5 paths.sh exit-vs-return: documented the "return 1 2>/dev/null || exit 1" pattern with a four-line comment. danielmiessler#7 Test gap closed: new rows in PAI-Install/engine/paths.test.ts cover cliClaude + PAI_DIR=whitespace and cliClaude + PAI_DIR=empty. (Was implicit before; now explicit.) 41 tests total (was 38). danielmiessler#6 Untracked dirs: added .claude/worktrees/ and .omc/ to superproject gitignore; added .omc/ to release-tree gitignore. Architecture references: switched from "L18-34" to '"## Directory Structure"' (section name, stable across edits) in hooks/lib/paths.ts, PAI/lib/paths.ts, PAI/lib/paths.sh, PAI/lib/paths.bats, hooks/lib/paths.test.ts, PAI/PAI-Install/engine/paths.ts, PAI/PAI-Install/engine/paths.test.ts, all 3 task YAMLs and the suite YAML. Old eval Results/ JSONs left as historical run artifacts. Verification: 41/41 TS GREEN (17 installer + 24 path-lib including new drift guard); 12/12 bats GREEN; shell helper smoke clean. Not addressed in this commit (per review): - #1 rebase onto origin/main (your call to push) - #2 settings.json/CLAUDE.md local overrides (left uncommitted)
neilinger
added a commit
to neilinger/Personal_AI_Infrastructure
that referenced
this pull request
May 3, 2026
Addresses code-review findings post-installer-refactor: danielmiessler#3 Mirror-lib drift guard: hooks/lib/paths.test.ts adds a unit test that diffs hooks/lib/paths.ts against PAI/lib/paths.ts (post-docblock). The two are deliberate mirrors per C1; this test fails the suite if they drift. No CI plumbing needed — runs in the regular bun test pass. danielmiessler#4 engine/paths.ts:53 clean() consistency: replaced direct process.env.PAI_DIR.trim() check with a clean() helper invocation matching the rest of the function. Same semantics, explicit intent, no future drift if clean() ever changes. danielmiessler#5 paths.sh exit-vs-return: documented the "return 1 2>/dev/null || exit 1" pattern with a four-line comment. danielmiessler#7 Test gap closed: new rows in PAI-Install/engine/paths.test.ts cover cliClaude + PAI_DIR=whitespace and cliClaude + PAI_DIR=empty. (Was implicit before; now explicit.) 41 tests total (was 38). danielmiessler#6 Untracked dirs: added .claude/worktrees/ and .omc/ to superproject gitignore; added .omc/ to release-tree gitignore. Architecture references: switched from "L18-34" to '"## Directory Structure"' (section name, stable across edits) in hooks/lib/paths.ts, PAI/lib/paths.ts, PAI/lib/paths.sh, PAI/lib/paths.bats, hooks/lib/paths.test.ts, PAI/PAI-Install/engine/paths.ts, PAI/PAI-Install/engine/paths.test.ts, all 3 task YAMLs and the suite YAML. Old eval Results/ JSONs left as historical run artifacts. Verification: 41/41 TS GREEN (17 installer + 24 path-lib including new drift guard); 12/12 bats GREEN; shell helper smoke clean. Not addressed in this commit (per review): - #1 rebase onto origin/main (your call to push) - #2 settings.json/CLAUDE.md local overrides (left uncommitted)
neilinger
added a commit
to neilinger/Personal_AI_Infrastructure
that referenced
this pull request
May 3, 2026
Addresses code-review findings post-installer-refactor: danielmiessler#3 Mirror-lib drift guard: hooks/lib/paths.test.ts adds a unit test that diffs hooks/lib/paths.ts against PAI/lib/paths.ts (post-docblock). The two are deliberate mirrors per C1; this test fails the suite if they drift. No CI plumbing needed — runs in the regular bun test pass. danielmiessler#4 engine/paths.ts:53 clean() consistency: replaced direct process.env.PAI_DIR.trim() check with a clean() helper invocation matching the rest of the function. Same semantics, explicit intent, no future drift if clean() ever changes. danielmiessler#5 paths.sh exit-vs-return: documented the "return 1 2>/dev/null || exit 1" pattern with a four-line comment. danielmiessler#7 Test gap closed: new rows in PAI-Install/engine/paths.test.ts cover cliClaude + PAI_DIR=whitespace and cliClaude + PAI_DIR=empty. (Was implicit before; now explicit.) 41 tests total (was 38). danielmiessler#6 Untracked dirs: added .claude/worktrees/ and .omc/ to superproject gitignore; added .omc/ to release-tree gitignore. Architecture references: switched from "L18-34" to '"## Directory Structure"' (section name, stable across edits) in hooks/lib/paths.ts, PAI/lib/paths.ts, PAI/lib/paths.sh, PAI/lib/paths.bats, hooks/lib/paths.test.ts, PAI/PAI-Install/engine/paths.ts, PAI/PAI-Install/engine/paths.test.ts, all 3 task YAMLs and the suite YAML. Old eval Results/ JSONs left as historical run artifacts. Verification: 41/41 TS GREEN (17 installer + 24 path-lib including new drift guard); 12/12 bats GREEN; shell helper smoke clean. Not addressed in this commit (per review): - #1 rebase onto origin/main (your call to push) - #2 settings.json/CLAUDE.md local overrides (left uncommitted)
konanzin
pushed a commit
to konanzin/Personal_AI_Infrastructure
that referenced
this pull request
May 20, 2026
… recovery Implementa handler completo para quando sessões terminam com erro: 1. Relatório estruturado salvo em MEMORY/LEARNING/SYSTEM/errors/: - Tipo, mensagem, stack trace do erro - Contexto da sessão (phase, mode, arquivos modificados) - Status de tentativa de recovery 2. Backup de emergência do work-in-progress: - Salva current-work.json em STATE/emergency/ - Permite recuperar trabalho se sessão crashar 3. Contadores de erros: - error-summary.json com total e por tipo - Acompanhamento semanal de erros - Histórico: erros por tipo e semana 4. Atualização do work.json: - Marca sessão com hasError: true - Registra errorType e errorTimestamp - Validação: 84/84 checks passando (100%) - Versão do plugin: 2.6.0 Refs: BACKLOG-HOOKS-PAI.md danielmiessler#6
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
/Users/daniel/with tilde shorthand~/in Claude settingsChanges
.claude/settings.jsonto use~/instead of hardcoded user directory🤖 Generated with Claude Code