### Terms and Conditions
- [x] I agree to the [Grant Agreement](https://9ba4718…c-5c73-47c3-a024-4fc4e5278803.usrfiles.com/ugd/9ba471_f81ef4e4b5f040038350270590eb2e42.pdf) terms if funded
- [x] I agree to [Provide KYC information](https://9ba4718c-5c73-47c3-a024-4fc4e5278803.usrfiles.com/ugd/9ba471_7d9e73d16b584a61bae92282b208efc4.pdf) if funded above $50,000 USD
- [x] I agree to disclose conflicts of interest
- [x] I agree to adhere to the [Code of Conduct](https://forum.zcashcommunity.com/t/zcg-code-of-conduct/41787) and [Communication Guidelines](https://forum.zcashcommunity.com/t/zcg-communication-guidelines/44284)
- [x] I understand all milestone deliverables will be validated and accepted by their intended users or their representatives, who will confirm that the deliverables meet the required quality, functionality, and usability for each user story.
- [x] I agree that for any new open-source software, I will create a `CONTRIBUTING.md` file that reflects the high standards of Zcash development, using the [`librustzcash` style guides](https://github.com/zcash/librustzcash/blob/main/CONTRIBUTING.md#styleguides) as a primary reference.
- [x] I understand when contributing to existing Zcash code, I am required to adhere to the project specific contribution guidelines, paying close attention to any [merge](https://github.com/zcash/librustzcash/blob/main/CONTRIBUTING.md#merge-workflow), [branch](https://github.com/zcash/librustzcash/blob/main/CONTRIBUTING.md#branch-history), [pull request](https://github.com/zcash/librustzcash/blob/main/CONTRIBUTING.md#pull-request-review), and [commit](https://github.com/zcash/librustzcash/blob/main/CONTRIBUTING.md#commit-messages) guidelines as exemplified in the `librustzcash` repository.
- [x] I agree to post request details on the [Community Forum](https://forum.zcashcommunity.com/c/grants/33)
- [x] I understand it is my responsibility to post a link to this issue on the [Zcash Community Forums](https://forum.zcashcommunity.com/c/grants/33) after this application has been submitted so the community can give input. I understand this is required in order for ZCG to discuss and vote on this grant application.
### Application Owners (@Octocat, @Octocat1)
@manfromnowhere143
### Organization Name
Aweb Labs
### How did you learn about Zcash Community Grants
Through the Zcash Foundation grants page, Zcash Community Grants site, and Zcash Grants Hub while looking for privacy-focused public-good funding for open-source agent execution infrastructure.
### Requested Grant Amount (USD)
$48,000
### Category
Infrastructure
### Project Lead
```project-lead.yaml
Name: Daniel Wahnich
Role: Founder and project lead, Aweb Labs
Background: Daniel is building Aweb as a governed execution layer for AI agents across models, MCP tools, APIs, cloud services, providers, and background workers. The public grant wedge is Aweb Agent Receipts: open-source execution evidence that records scoped authority, tool/provider execution, outcome, cost, failure, retry, recovery state, and redaction boundaries.
Responsibilities: Product and technical direction, Zcash-aligned receipt profile, TypeScript reference implementation, CLI validator, sample corpus, privacy/redaction documentation, lightweight viewer, community updates, and milestone delivery.
```
### Additional Team Members
```team-members.yaml
None or N/A. This is a founder-led milestone. If funded, limited external review or security/privacy feedback may be commissioned inside the milestone budget, but there are no additional core team members at submission time.
```
### Project Summary
Aweb Zcash Private Agent Receipts is an open-source privacy receipt layer for AI-agent execution. It records authority, provider/tool use, outcome, cost, failure, and recovery state while redacting prompts, secrets, counterparties, addresses, viewing keys, private notes, and raw workflow context.
### Project Description
AI agents are starting to call APIs, use MCP tools, route messages, operate cloud workers, and touch payment-adjacent workflows. The evidence of those actions is usually trapped in raw logs, provider dashboards, traces, prompts, inboxes, and private operator notes. That creates a hard tradeoff: either keep everything and create surveillance-grade execution records, or keep almost nothing and lose accountability.
Aweb Zcash Private Agent Receipts applies Zcash-aligned privacy discipline to agent execution evidence. The project will define a private-by-default receipt profile, redaction model, selective-disclosure schema, TypeScript helpers, CLI validator, sample corpus, documentation, and lightweight viewer. The goal is to preserve enough evidence for review, debugging, and public-good accountability without exposing raw prompts, secrets, addresses, counterparties, viewing keys, or private workflow context.
This is not a custody product, wallet, token launch, or claim of live Zcash payment execution. It is open-source developer infrastructure for privacy-preserving agent accountability, designed so future Zcash builders can reason about what an agent was allowed to do, what it actually did, what failed, what it cost, and what recovery state remains without leaking the private details that Zcash users expect to protect.
### Proposed Problem
Agent execution logs can become a privacy liability. A transaction hash or provider event is too narrow to explain an agent workflow, but raw logs often contain prompts, user data, API arguments, addresses, counterparties, memos, provider payloads, and private operator notes.
For privacy-preserving ecosystems like Zcash, this is especially important. Builders need a way to review agent behavior and failures without normalizing public raw logs. The missing primitive is a compact receipt that records scoped authority, public outcome, cost, failure, retry, recovery, and redaction decisions while keeping private material private.
### Proposed Solution
Aweb will build a Zcash-aligned Agent Receipts profile for private execution evidence.
The receipt records:
1. What authority the agent had before acting.
2. Which capability, provider, API, MCP tool, or workflow step was used.
3. What happened, including status, cost, failure, retry, and recovery state.
4. Which data was intentionally withheld or redacted.
5. Optional digest or reference fields for public review without publishing private payloads.
6. Guidance for optional Zcash transaction-reference fields without exposing addresses, viewing keys, memos, or counterparties.
The first funded milestone will produce the profile, TypeScript helpers, JSON Schema, CLI validator, sample receipts, documentation, lightweight viewer, and final report. It will demonstrate the privacy boundary with simulated fixtures first, then leave a clean path for future Zcash-specific integration work if the community wants it.
### Solution Format
Open-source software and documentation:
1. Zcash-aligned Agent Receipt profile and redaction taxonomy.
2. TypeScript reference helpers and JSON Schema.
3. CLI validator for required fields, redaction checks, and sample validation.
4. Sample receipt corpus for success, failure, rejection, retry, recovery, and payment-adjacent private workflows.
5. Lightweight receipt viewer.
6. Integration documentation and final report for Zcash builders.
The deliverables will be published under Apache-2.0 unless ZCG requests a different compatible open-source license.
### Dependencies
Technical dependencies:
- Node.js / TypeScript tooling.
- JSON Schema validation.
- Existing Aweb Agent Receipts examples and schema work.
- GitHub for open-source publication, issue tracking, and ZCG submission.
- Public documentation hosting.
Zcash dependencies:
- Zcash Community Grants feedback.
- Zcash Community Forum posting and review.
- Review of Zcash privacy expectations around addresses, memos, viewing keys, transaction references, and selective disclosure.
The milestone does not depend on custody, token issuance, live user funds, or production shielded payments.
### Technical Approach
1. Define the Zcash privacy receipt profile:
- public fields,
- private/redacted fields,
- redaction reasons,
- disclosure scopes,
- consent and counterparty boundaries,
- optional Zcash transaction-reference guidance.
2. Implement TypeScript helpers and schema validation:
- create receipts,
- validate required fields,
- redact unsafe fields,
- verify no raw prompt, secret, address, memo, viewing key, or private provider payload leaks into public fixtures.
3. Build a CLI validator:
- validate sample corpus,
- print missing fields,
- fail on unsafe public fields,
- produce reviewer-friendly output.
4. Build examples:
- simulated success,
- failed/rejected action,
- retry/recovery,
- payment-adjacent private workflow,
- human-review path,
- redaction report.
5. Build lightweight viewer:
- show authority, action, provider/tool, status, failure, recovery, cost, and privacy boundary,
- hide or label private/redacted data.
6. Publish docs and final report:
- README,
- contribution guide,
- redaction guide,
- Zcash alignment note,
- final continuation recommendation.
### Upstream Merge Opportunities
This proposal does not plan to fork or modify existing Zcash protocol software during the first milestone. It avoids direct changes to zcashd, Zebra, librustzcash, wallet code, or consensus-critical repositories.
Upstream opportunities are documentation and ecosystem-tooling oriented:
- Zcash builders may reuse the receipt profile and redaction taxonomy.
- Future wallet, SDK, or infrastructure teams could adapt the receipt fields for private workflow review.
- If the community wants deeper integration later, the final report will identify whether the next step should be Zcash wallet/SDK examples, transaction-reference guidance, or privacy-preserving application integration.
No upstream merge is required for the first milestone to be successful.
### Hardware/Software Costs (USD)
$3,000
### Hardware/Software Justification
Development and testing tools, domains/hosting, CI, package publishing, local test infrastructure, and documentation/viewer deployment needed to build and validate the open-source receipt profile and examples.
### Service Costs (USD)
$5,000
### Service Costs Justification
External review reserve for privacy/security feedback, documentation review, and community-driven corrections after ZCG and forum feedback. This is not for marketing spend or token promotion.
### Compensation Costs (USD)
$40,000
### Compensation Costs Justification
Founder/developer compensation for approximately five months of focused design, implementation, testing, documentation, community updates, revisions, and final reporting. Work includes the receipt profile, TypeScript helpers, CLI validator, sample corpus, viewer, docs, and forum/community feedback cycle.
### Total Budget (USD)
$48,000
### Previous Funding
No
### Previous Funding Details
Aweb has not received previous Zcash Community Grants funding.
### Other Funding Sources
Yes
### Other Funding Sources Details
Aweb Labs is founder-funded with a small personal seed investment. Other infrastructure support or startup credits reported in the broader Aweb build include Neon credits, E2B credits, Mixpanel for Startups, and Amplitude Startup Scholarship. These do not fund the Zcash-specific Agent Receipts milestone and do not replace this grant request.
### Implementation Risks
Main risks:
1. The receipt profile could become too broad for practical adoption.
2. Privacy boundaries could be under-specified.
3. Builders may disagree about which facts should be public, digest-only, or private.
4. The first milestone may need revisions after ZCG and community feedback.
Mitigation:
Keep the first milestone narrow, publish a concrete schema and examples, validate every sample, document explicit redaction rules, avoid custody or live-funds claims, and use the Zcash forum review process to improve the privacy boundary before deeper integration.
### Potential Side Effects
Potential negative side effect: bad receipt design could encourage over-publication of sensitive workflow metadata.
Mitigation: the proposal is private-by-default. It explicitly blocks raw prompts, secrets, counterparties, addresses, viewing keys, memos, raw provider payloads, and private operator notes from public examples. The public surface is limited to scoped authority, status, failure, recovery, cost, redaction reason, and optional digest/reference fields.
The project will not move user funds, create custody flows, publish viewing keys, or claim production Zcash payment execution in the first milestone.
### Success Metrics
1. Public Zcash-aligned receipt profile published.
2. TypeScript helpers and JSON Schema published under Apache-2.0.
3. CLI validator passes the sample corpus.
4. At least 8 sample receipts cover success, failure, rejection, retry, recovery, human review, redaction report, and payment-adjacent private workflow.
5. No sample contains raw prompts, secrets, addresses, counterparties, viewing keys, memos, private tool arguments, or provider credentials.
6. Lightweight viewer renders authority, tool/provider, privacy boundary, outcome, failure, retry, recovery, and cost.
7. Documentation includes a redaction guide, contribution guide, Zcash alignment note, and final report.
8. Zcash community reviewers can inspect the proposal and understand how accountability is preserved without surveillance-grade raw logs.
### Startup Funding (USD)
$0
### Startup Funding Justification
No separate startup funding is requested. The full $48,000 request is allocated through the milestones below, so startup funding plus milestone amounts equals the total grant request.
### Milestone Details
```milestones.yaml
- Milestone: 1
Amount (USD): 8000
Expected Completion Date: 2026-07-31
User Stories:
- "As a Zcash builder, I want a clear privacy receipt profile, so that I can understand what agent execution facts may be public and what must stay private."
- "As a reviewer, I want a redaction taxonomy, so that I can evaluate whether the project avoids surveillance-grade raw logs."
Deliverables:
- Zcash-aligned receipt profile.
- Threat model and redaction taxonomy.
- Public/private/digest-only field guidance.
- Zcash alignment note covering addresses, memos, viewing keys, counterparties, and optional transaction references.
Acceptance Criteria: Profile and redaction taxonomy are published; at least 5 receipt scenarios are specified; no live-funds or custody claims are made.
- Milestone: 2
Amount (USD): 13000
Expected Completion Date: 2026-08-31
User Stories:
- "As a developer, I want TypeScript helpers and schema validation, so that I can create privacy-aware agent receipts locally."
- "As a reviewer, I want the validator to fail unsafe fields, so that raw private data does not enter public examples."
Deliverables:
- TypeScript reference helpers.
- JSON Schema.
- CLI validator.
- Redaction checks for prompts, secrets, addresses, memos, viewing keys, private tool args, and provider credentials.
- Initial tests and usage examples.
Acceptance Criteria: CLI validates expected examples and fails unsafe fixtures; package can create and validate Zcash-aligned private receipts locally.
- Milestone: 3
Amount (USD): 9000
Expected Completion Date: 2026-09-30
User Stories:
- "As a Zcash community reviewer, I want realistic sample receipts, so that I can inspect how the privacy boundary works."
- "As a privacy-preserving app builder, I want examples for failure and recovery, so that I can model non-happy-path workflows without leaking private data."
Deliverables:
- Sample receipt corpus for success, failure, rejection, retry, recovery, human review, redaction report, and payment-adjacent private workflow.
- Example documentation for each scenario.
- Validation report for all sample receipts.
Acceptance Criteria: At least 8 sample receipts pass validation; no raw private data appears in public fixtures; each example explains disclosed and withheld fields.
- Milestone: 4
Amount (USD): 10000
Expected Completion Date: 2026-10-31
User Stories:
- "As a human reviewer, I want a lightweight viewer, so that I can inspect authority, status, privacy boundary, failure, and recovery without reading raw JSON."
- "As an integrator, I want documentation, so that I can reuse the profile in privacy-preserving agent workflows."
Deliverables:
- Lightweight receipt viewer.
- Public README.
- Integration guide.
- Redaction guide.
- Contribution guide with Zcash-style open-source standards.
Acceptance Criteria: Viewer renders receipt authority/tool/privacy/outcome/failure/recovery/cost fields; docs explain how to create, validate, redact, and review receipts.
- Milestone: 5
Amount (USD): 8000
Expected Completion Date: 2026-11-30
User Stories:
- "As ZCG and community reviewers, we want a final report and corrections, so that the project reflects community feedback."
- "As future builders, we want a continuation recommendation, so that the next step is clear and scoped."
Deliverables:
- Community feedback pass.
- Corrections and additional tests.
- External/security/privacy review reserve output if used.
- Final report.
- Recommendation for future Zcash wallet/SDK/app integration path.
Acceptance Criteria: Final report is published; open issues from review are addressed or documented; continuation path is clearly recommended.
```
### Supporting Documents
```files.yaml
- Link 1: https://aweb-wine.vercel.app/grants/agent-receipts - Public reviewer dossier for Aweb Agent Receipts, including the open-source execution-evidence scope and supporting context for the Zcash privacy receipt proposal.
- Link 2: https://aweb-wine.vercel.app/specs/aweb-agent-receipts.md - Public technical specification for the Aweb Agent Receipts format and validation direction.
- Link 3: https://github.com/manfromnowhere143 - Applicant GitHub profile. The funded Zcash-specific package would be extracted into a public Apache-2.0 repository/package if approved.
- File prepared for upload if requested: zcash-aweb-fixed.pdf - Fixed three-page Aweb Zcash Private Agent Receipts proposal packet containing project framing, deliverables, budget, milestones, success metrics, and public reviewer links.
```
- Attached PDF: [zcash-aweb-fixed.pdf](https://github.com/user-attachments/files/28328192/zcash-aweb-fixed.pdf)