diff --git a/.github/instructions/dictation.instructions.md b/.github/instructions/dictation.instructions.md index 02c963f1e6a..2126b8ae429 100644 --- a/.github/instructions/dictation.instructions.md +++ b/.github/instructions/dictation.instructions.md @@ -54,7 +54,6 @@ lockfile logger logs main -makefile max-concurrency max-turns mcp @@ -92,7 +91,6 @@ session smoke source staged -starlight step steps stop-after diff --git a/.github/workflows/dictation-prompt.lock.yml b/.github/workflows/dictation-prompt.lock.yml index 112e8c4a21f..7325a42e284 100644 --- a/.github/workflows/dictation-prompt.lock.yml +++ b/.github/workflows/dictation-prompt.lock.yml @@ -1024,13 +1024,13 @@ jobs: ```bash # Scan documentation markdown files for technical terms - find docs -name "*.md" -type f -exec grep -hoE '\b[a-z][a-z0-9_-]{2,}\b' {} + | sort -u | head -200 + find docs/src/content/docs -name "*.md" -type f -exec grep -hoE '\b[a-z][a-z0-9_-]{2,}\b' {} + | sort -u | head -200 # Extract headings from documentation - find docs -name "*.md" -type f -exec grep -h "^#" {} + | sed 's/^#* //' | head -100 + find docs/src/content/docs -name "*.md" -type f -exec grep -h "^#" {} + | sed 's/^#* //' | head -100 # Check key documentation files - cat docs/README.md docs/src/content/docs/reference/*.md | grep -oE '\b[a-z][a-z0-9_-]{3,}\b' | sort -u | head -150 + cat docs/src/content/docs/reference/*.md | grep -oE '\b[a-z][a-z0-9_-]{3,}\b' | sort -u | head -150 ``` **Extract terms from documentation covering:** @@ -1046,6 +1046,8 @@ jobs: **Goal**: Create a list of approximately 100 terms (95-105 is acceptable) that are most relevant to this project, extracted from documentation only. + **Exclude tooling-specific terms**: Do not include makefile, Astro, or starlight as these are tooling-specific and not user-facing concepts. + ### 2. Create the Dictation Instructions File Create the file `.github/instructions/dictation.instructions.md` with the following structure (NO EXAMPLES): @@ -1122,7 +1124,8 @@ jobs: ## Important Guidelines - - **Extract from documentation**: Scan only docs/**.md files for terms + - **Extract from documentation**: Scan only docs/src/content/docs/**/*.md files for terms + - **Ignore tooling-specific terms**: Exclude makefile, Astro, and starlight (tooling-specific, not user-facing) - **Be Thorough**: Scan all documentation files to extract accurate terms - **Be Precise**: The glossary should contain approximately 100 terms (95-105 is acceptable) - **Prioritize Relevance**: Include terms that are actually used frequently in documentation @@ -1137,7 +1140,8 @@ jobs: Your task is complete when: 1. ✅ The file `.github/instructions/dictation.instructions.md` exists 2. ✅ It contains approximately 100 project-specific terms in the glossary (95-105 is acceptable) - 3. ✅ Terms are extracted from documentation files only (docs/**.md) + 3. ✅ Terms are extracted from documentation files only (docs/src/content/docs/**/*.md) + 4. ✅ Tooling-specific terms (makefile, Astro, starlight) are excluded 4. ✅ The rephrasing guidelines are clear and actionable 5. ✅ NO examples are included (keep it short) 6. ✅ File description removed and rephrased as a prompt diff --git a/.github/workflows/dictation-prompt.md b/.github/workflows/dictation-prompt.md index 6f827cbadae..ad03c8867c3 100644 --- a/.github/workflows/dictation-prompt.md +++ b/.github/workflows/dictation-prompt.md @@ -53,13 +53,13 @@ Extract a comprehensive glossary of terms specific to this project from document ```bash # Scan documentation markdown files for technical terms -find docs -name "*.md" -type f -exec grep -hoE '\b[a-z][a-z0-9_-]{2,}\b' {} + | sort -u | head -200 +find docs/src/content/docs -name "*.md" -type f -exec grep -hoE '\b[a-z][a-z0-9_-]{2,}\b' {} + | sort -u | head -200 # Extract headings from documentation -find docs -name "*.md" -type f -exec grep -h "^#" {} + | sed 's/^#* //' | head -100 +find docs/src/content/docs -name "*.md" -type f -exec grep -h "^#" {} + | sed 's/^#* //' | head -100 # Check key documentation files -cat docs/README.md docs/src/content/docs/reference/*.md | grep -oE '\b[a-z][a-z0-9_-]{3,}\b' | sort -u | head -150 +cat docs/src/content/docs/reference/*.md | grep -oE '\b[a-z][a-z0-9_-]{3,}\b' | sort -u | head -150 ``` **Extract terms from documentation covering:** @@ -75,6 +75,8 @@ cat docs/README.md docs/src/content/docs/reference/*.md | grep -oE '\b[a-z][a-z0 **Goal**: Create a list of approximately 100 terms (95-105 is acceptable) that are most relevant to this project, extracted from documentation only. +**Exclude tooling-specific terms**: Do not include makefile, Astro, or starlight as these are tooling-specific and not user-facing concepts. + ### 2. Create the Dictation Instructions File Create the file `.github/instructions/dictation.instructions.md` with the following structure (NO EXAMPLES): @@ -151,7 +153,8 @@ After creating the dictation instructions file: ## Important Guidelines -- **Extract from documentation**: Scan only docs/**.md files for terms +- **Extract from documentation**: Scan only docs/src/content/docs/**/*.md files for terms +- **Ignore tooling-specific terms**: Exclude makefile, Astro, and starlight (tooling-specific, not user-facing) - **Be Thorough**: Scan all documentation files to extract accurate terms - **Be Precise**: The glossary should contain approximately 100 terms (95-105 is acceptable) - **Prioritize Relevance**: Include terms that are actually used frequently in documentation @@ -166,7 +169,8 @@ After creating the dictation instructions file: Your task is complete when: 1. ✅ The file `.github/instructions/dictation.instructions.md` exists 2. ✅ It contains approximately 100 project-specific terms in the glossary (95-105 is acceptable) -3. ✅ Terms are extracted from documentation files only (docs/**.md) +3. ✅ Terms are extracted from documentation files only (docs/src/content/docs/**/*.md) +4. ✅ Tooling-specific terms (makefile, Astro, starlight) are excluded 4. ✅ The rephrasing guidelines are clear and actionable 5. ✅ NO examples are included (keep it short) 6. ✅ File description removed and rephrased as a prompt