You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: en/ai-sre/knowledge.mdx
+5-5Lines changed: 5 additions & 5 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -32,7 +32,7 @@ Knowledge Packs are one type of AI SRE resource and follow the same two-level sc
32
32
33
33
`DUTY.md` is the **table of contents entry point** for the entire Knowledge Pack. It is the catalog itself — the agent reads `DUTY.md` in full, then fetches other files on demand via `@filename` references. The system does not attach a separate file listing alongside `DUTY.md`; the catalog is the body.
34
34
35
-
References use the `@<filename>` style, where the filename points to another file within the same pack (flat naming, no directory hierarchy):
35
+
References use the `@<path>` style, where the path points to another file within the same pack. Subdirectories are supported (e.g., `@runbooks/api-5xx.md`):
36
36
37
37
```markdown
38
38
# On-call Knowledge Overview (DUTY.md)
@@ -41,8 +41,8 @@ References use the `@<filename>` style, where the filename points to another fil
41
41
Our core services and owners: @services.md.
42
42
43
43
## Common Incident Handling
44
-
- API 5xx spike: see @runbook-api-5xx.md
45
-
- Database connection pool exhaustion: see @runbook-db-pool.md
44
+
- API 5xx spike: see @runbooks/api-5xx.md
45
+
- Database connection pool exhaustion: see @runbooks/db-pool.md
46
46
47
47
## Clusters & Environments
48
48
Production cluster topology and access: @cluster.yaml.
@@ -62,7 +62,7 @@ After reading `DUTY.md`, the agent decides which `@references` to expand based o
62
62
| Per-file size limit | 1 MiB | Files exceeding this cannot be saved |
63
63
| Per-pack size limit | 5 MiB | The usage bar in the console reflects this quota |
64
64
| File count limit | 100 | No new files can be added once the limit is reached |
65
-
|Directory hierarchy | Not allowed | Filenames must be flat — no `/` or subdirectories|
65
+
|Subdirectories | Allowed | Paths may contain `/` (e.g., `runbooks/api-5xx.md`); path segments cannot start with `.`|
66
66
| Dot files | Not allowed | Filenames cannot start with `.`|
67
67
68
68
## Create & Edit
@@ -146,7 +146,7 @@ This two-level scope model applies uniformly to all resources within the account
146
146
Keep `DUTY.md` to link lists and one-line descriptions only — all substantive content goes into the sibling files it `@references`. This keeps the catalog concise and readable, and lets the agent expand only the branches relevant to the current incident, avoiding irrelevant content consuming context.
147
147
</Accordion>
148
148
<Accordiontitle="One topic per file"icon="wrench">
149
-
Focus each runbook on one incident type or one service (e.g., `runbook-api-5xx.md`, `runbook-db-pool.md`), using clear filenames as semantic indexes. Since packs use flat naming (no subdirectories), filename prefixes are your "directory structure."
149
+
Focus each runbook on one incident type or one service (e.g., `runbooks/api-5xx.md`, `runbooks/db-pool.md`), using clear paths as semantic indexes. Subdirectories are supported — you can group files by service or topic (e.g., `runbooks/`, `configs/`).
Service catalogs, cluster topologies, and threshold configs are well-suited to `.yaml` / `.json` (e.g., `services.md`, `cluster.yaml`), so the agent can both read and parse them directly. Script snippets can use `.sh`.
|`MIRROR_URL`| Override the release asset download mirror prefix used by the installer script; must use `https://`|`https://static.flashcat.cloud/flashduty-cli`|
44
+
|`FLASHDUTY_UPDATE_BASE_URL`| Override the base URL used by `flashduty update` and the automatic update check |`https://static.flashcat.cloud/flashduty-cli`|
0 commit comments