Add optional `escapeJson` parameter to `ConfigPaths.substitute` to control
whether file content should be JSON-escaped. When `false`, multiline content
and special characters are preserved as-is, which is required for agent
markdown prompts that need raw file content rather than escaped strings.
Update agent config processing to pass `escapeJson: false` when substituting
file content into markdown prompts, ensuring proper handling of multiline
files without unwanted escape sequences.
Add comprehensive test coverage for the new parameter and existing
substitution functionality.
Enable file content substitution in agent configuration prompts using the
existing `ConfigPaths.substitute` utility. This allows agent markdown files
to reference external file contents via `{file:path}` syntax, with missing
files resolving to empty strings.
Export `substitute` function from ConfigPaths to support agent config
processing.
* fix: catch EEXIST from recursive mkdir on Windows
fs.mkdir with { recursive: true } can still throw EEXIST on Windows
when the target directory is backed by an NTFS reparse point (OneDrive),
a directory junction, or a WSL-served path. libuv's recursive walk
doesn't treat these as regular directories, so mkdir with recursive:true
attempts to create an already-existing path segment and fails.
This adds a defensive EEXIST catch at both directory-creation origins:
- AppFileSystem.ensureDir / writeWithDirs (snapshots, config, plans)
- Encoding.write (all tool writes and edits)
No functional change on Linux/macOS where EEXIST is never thrown from
recursive mkdir.
Fixes#9618, #9755
* fix: remove extra arg from mkdirSafe call, add unit tests
* docs(kilo-docs): update auto-generated source links
Add 2 new GitHub issue references (9618, 9755) from encoding.ts
and update URL count from 84 to 86.
---------
Co-authored-by: nimgrim <nimgrim@users.noreply.github.com>
Co-authored-by: Imanol Maiztegui <imanol.mzd@gmail.com>
\S+ was greedily consuming trailing punctuation (. ! , ;) into the URL
match, causing new URL() to mutate clean ASCII URLs — e.g.
"see https://example.com." became "see https://example.com./".
Fix: strip trailing sentence punctuation before parsing, extract and
replace only the hostname (not href) to avoid adding trailing slashes to
bare origins, and restore the stripped tail afterward.
* fix(tool): clarify semantic_search returns snippets not file paths
The first description bullet read "Find files most relevant to the
search query using semantic search.", which led an AI agent to tell a
user that the tool returns file paths only. The tool actually returns
matching content with file paths, line ranges, and relevance scores.
Updates the first two description bullets to name the output shape
directly. No runtime change.
Closes#9767.
* test(cli): cover semantic_search description
---------
Co-authored-by: marius-kilocode <marius@kilocode.ai>
The existing apply_patch encoding test only checked final file bytes,
which were correct because Patch.deriveNewContentsFromChunks does its
own encoding-aware read. The diff and additions/deletions counts
returned to the user (and shown in the permission prompt) were
nonetheless garbled because the surrounding tool reused a hard-coded
UTF-8 decoder. Pin the diff metadata to catch that regression.
Kilo isn't always BYOK; what matters here is that Kilo doesn't charge a
per-request fee on top of the upstream inference cost. So for the Kilo
provider, the gateway/marketplace `cost` would understate the user's true
spend — always prefer the upstream/market cost when present.
* feat: support Kilo-hosted indexing embeddings
* fix: format indexing settings
* fix: serialize local model persistence
* fix: annotate local model persistence
* fix: stabilize MCP OAuth browser test
* fix: annotate MCP OAuth test wait
* fix: keep KiloProvider under line cap
* fix: align Kilo indexing auth defaults
* fix(vscode): add missing indexing translations
* feat(indexing): fetch Kilo embedding model catalog
* fix(indexing): align Kilo embedding metadata fallback
* fix(gateway): remove unused indexing dependency
* fix(indexing): keep fallback catalog webview safe
* fix(agent-manager): include embedding model provider
* fix(vscode): provide embedding models in stories
* refactor(indexing): rely on Cloud embedding catalog
* fix(indexing): hydrate Kilo model metadata from Cloud
* fix(indexing): clear model when switching provider
Move the new Anthropic Messages, OpenAI Responses and Vercel AI Gateway
cost-extraction tests out of the shared upstream `test/session/compaction.test.ts`
file into `test/kilocode/provider-cost.test.ts` to keep Kilo-specific test
coverage in a Kilo-owned location and minimize merge conflicts with upstream.
Refactor providerCost to read `usage.raw.cost_details.upstream_inference_cost`,
which both `@ai-sdk/anthropic` (`convertAnthropicUsage`) and `@ai-sdk/openai`
(`convertOpenAIResponsesUsage`) populate with the verbatim OpenRouter usage
object. This unifies the Anthropic Messages and OpenAI Responses cost paths
through a single branch and removes the redundant `metadata.anthropic.usage`
branch (the SDK exposes the same data under both keys for Anthropic, and only
under `raw` for Responses).
Move KILO_CONFIG_DIR profile fallback tests out of the shared upstream
instruction.test.ts into packages/opencode/test/kilocode/instruction.test.ts
to reduce future merge conflicts with opencode.
For the Anthropic Messages API (via OpenRouter or Vercel AI Gateway), the
top-level `cost` field is the gateway fee that Kilo paid (typically 0 for
BYOK), not the user's true spend. Only use `upstream_inference_cost` /
`marketCost`; fall back to the token-based calculation when those are
absent.
Cost reporting already worked when Kilo used OpenRouter chat completions
internally. Extend providerCost to also read cost from the Anthropic
Messages API stream metadata, covering both OpenRouter (Anthropic-style
`usage.cost` / `cost_details.upstream_inference_cost`) and Vercel AI
Gateway (`gateway.cost` / `gateway.marketCost`).
marius-kilocode